The Everyday Making of Citizen Journalism on Twitter

To Yell @ the Wind: The Everyday Making of Citizen Journalism on Twitter

June 16, 2013

Student: Supervisor: Diana Chichifoi Prof. Dr. Kristina Riegert

Master Thesis Media and Communication Studies Department of Media Studies (IMS) Stockholm University Abstract Manifestations of citizen journalism on Twitter have been mainly analyzed around trending stories, associated with public campaigns or breaking news. Microbloggers’ everyday attempts to contribute to professional media coverage, however, have been scarcely addressed in the literature. This research analyzes the practice of tweeting at inﬂuential news outlets by using a mixed-method approach. Content analysis and directed queries about users’ own motivation to tweet at media explore the interaction between regular users and professional news networks. Results show that there is a predisposition towards negative media commentary followed closely by newsworthy topics generated by the users, which replicate to a large extent mainstream media’s agenda. When asked directly in the feed ‘why do you tweet at media?’, users explain their motivations to address news outlets in diverging ways, from explicit citizen journalism mission, to visibility claims and coincidental tagging. The study contributes to the study of citizen journalism on social media platforms and elaborates a mixed method approach suited for this type of online data.

Keywords: citizen journalism, gatekeeping, agenda-setting, news values, content analysis, social media, Twitter

Acknowledgements I would like to thank Prof. Dr. Kristina Riegert for having taken the time and eﬀort to guide me through this academic journey of many a treacherous turn. I am grateful to my parents for inspiring and motivating me and for watching my steps so far; to my sister, for her help with coding, encouragements and open heart; to C˜alin for reviewing my work, for his patience and unconditional support.

2 Contents

1 Introduction 5

2 Materials and Research Aims 7

3 Literature Review 8 3.1 Twitter: features and premises of generating news ...... 8 3.2 Citizen journalism ...... 11 3.2.1 Citizen journalism on Twitter – Premises ...... 14 3.2.2 Citizen journalism on Twitter – Implications ...... 15 3.2.3 Gatekeeping ...... 15 3.2.4 Newsworthiness on Twitter ...... 17 3.2.5 Agenda-setting ...... 18 3.2.6 Some Broader Implications of Citizen Journalism on Twitter . . . . . 19 3.3 Self-promotion on Twitter ...... 21

4 Research design 24 4.1 A mixed-method approach ...... 24 4.2 Content analysis ...... 25 4.2.1 Sampling and data collection ...... 26 4.2.2 Operationalization of variables ...... 27 4.2.3 Pilot study ...... 28 4.2.4 Coding ...... 29 4.3 Qualitative interrogation – Twitter replies ...... 29 4.3.1 Operationalization ...... 30 4.3.2 Data collection ...... 31 4.4 Research evaluation ...... 32 4.4.1 Reliability ...... 33 4.4.2 Validity ...... 33 4.4.3 Replicability ...... 33

5 Results 34 5.1 Content Analysis – Results ...... 34 5.2 Twitter replies – Results ...... 37 5.3 An integrative perspective ...... 42

6 Discussion 43 6.1 Tweeting at news media – Purpose and dynamics ...... 43 6.2 What makes news according to twitterers ...... 47 6.3 Personal drive to yell @ the wind ...... 49

7 Conclusions 51

8 References 55

3 9 Appendix I 65 9.1 Codebook ...... 65 9.2 Decision rules ...... 73

10 Appendix II 74 10.1 Intercoder reliability tests ...... 74

11 Appendix III 75 11.1 Descriptive statistics ...... 75

12 Appendix IV 78 12.1 Collected Twitter replies ...... 78

4 1 Introduction

Playing the journalist, the media critic or the active citizen in a 140-character script is a diﬃcult mission. A one-time shout on Twitter that speaks a relevant message to the masses can raise global awareness and catch the media’s attention – or it could get lost in the feed. Since Twitter’s launch in 2006, users have developed practices of gathering and spreading the news using microblogging tools. When twitterers want to enrich their social media experience with a journalistic feel, they often participate in news production next to, or against, professional media. Tweeting ‘@’ inﬂuential media outlets happens daily and spontaneously, as users criticize, appraise or notify professional media about newsworthy content, even outside ‘trending topics’. Issues considered of public interest which are brought up by regular users stand for the everyday making of citizen journalism on Twitter. Messages directed at professional media are central to this study, which aims to explore the newsworthy content issued by twitterers and the users’ own reasons to interact with mainstream media on this platform.

The way information, and especially news, is collected and replicated using Twitter has been addressed in media research under various angles. Twitter features make it easy for users to filter content based on individual preferences and to discover trends with information value. What people ‘follow’, mention (@) or hashtag (#) in their messages reflects to a large extent their interests and Twitter routines. Users’ constant exposure to real-time news reporting also explains why the most popular, trending topics derive from news stories (Thelwall et al., 2011). As an interactive environment, Twitter not only affects “the way news is gathered, disseminated, and consumed” (Hermida, 2012, p. 2), but also encourages active responses to news. As a participatory tool, microblogging has mainly been analyzed around salient media events, such as social uprisings (Hermida, Lewis, and Zamith, 2012; Knight, 2012; Papacharissi & de Fatima Oliveira, 2012), political campaigns (Larsson & Moe, 2012; Christensen, 2013), natural disasters (Smith, 2010) crises (Levinson, 2009), or as source for breaking news (Murthy, 2011). In previous empirical studies content with maximum visibility in the network made the object of research, while the more personalized ways of interacting with media on Twitter have not been addressed.

News threads started on Twitter that became mainstream stories continue to inspire users to address their messages to professional media. Tagging the media in a tweet is a common way to reach out to a broader audience or more directly tell something to certain media outlets. This practice and the content thus generated have been scarcely

5 addressed in the literature so far. The reasons vary: such tweets emerge spontaneously from a personal motivation, do not have a common topic or purpose, and address different media outlets at a time, without a specific selection pattern. As a general practice, tweeting at professional media shows users’ awareness of how news is made and their active interest in influencing it. Providing an alternative to the mainstream news selection, a fact check or a critique, users’ tweeting agendas are an analytically interesting scenario which can describe the everyday performance of citizen journalism on Twitter. Similarities and differences between citizen-made journalism and professional news reporting can be explained based on this type of user-generated content. This research is motivated by the ‘everyday’ aspect of tweeting with a news agenda citizen journalism has ‘happened’ repeatedly on Twitter, and has received growing attention from media scholars. What has been missing, however, is an analysis of the daily exercises in news-making via Twitter. This study attempts to address the interaction between regular users and news media by establishing a typology of tweets and exploring the potential news agenda behind them.

Outline

The study begins with a general description of the research aims and materials in Chapter 2. The theoretical background in Chapter 3 presents the main directions of research that connect Twitter use with citizen journalism, the media’s gatekeeping and agenda-setting processes, and the more general social media routines. The research design in Chapter 4 ﬁrst outlines the mixed-method approach, with special consideration given to the adaptation of methods to the type of material and the online environment in general. The method description further progresses from content analysis to qualitative interrogation in the Twitter feed. The results of the two methods are presented in Chapter 5, following the three research questions. The implications of the ﬁndings are presented in Chapter 6, in the light of citizen journalism practices and social media use, with an eye on emerging directions of research. Chapter 7 concludes the study, by presenting potential methodological improvements and several questions raised by the present research.

6 2 Materials and Research Aims

The goal of this study is to explore messages and the intention behind the content directed by Twitter users at influential news networks. The analysis concentrates on tweets with multiple, i.e. two or more, news media addressees. The user-tagging function (i.e. ‘@username’) is used as a content selection tool, to define the population of units considered for analysis. A more detailed description of Twitter functions in the forthcoming chapter anticipates the roles assumed by the users when confronting professional media. The intention behind tweeting at media outlets can be analyzed under several angles. First, by looking strictly at the text content posted on Twitter, several reasons for addressing professional media can be mapped. Previous research on social media use, citizen journalism and participatory forms of news-making will provide the background to help identify these reasons. Second, users’ self-explained motivation to tweet at professional media can provide deeper insight into how personal microblogging habits are oriented by and around news media. In general, accounts tagged on Twitter are seen as direct recipients of external messages or as agents to promote a message further. Corporate accounts of news media outlets are technically set up as any other user’s, but are perceived as more influential because of their official mission to inform. That is why tagging news media has a potentially more complex background than @-mentioning just any other twitterer. The first research question tries to establish several directions of this online practice by looking at the content generated. RQ1: To what extent tweets addressed at professional media outlets represent newsworthy content, media commentary, personal statements and self-promotion material? These directions and associated types of content will be outlined based on social media practices and participatory forms of news-making in the literature review sections. The study further aims at exploring the content provided by users who tag media in their posts and concentrates on newsworthy stories thus generated. RQ2: What newsworthy topics are selected by users when tweeting at news media outlets? The last part of the analysis complements the broad classification of tweets with a qualitative thematic analysis. The aim is to explore users’ self-declared motivation to tweet at professional media. RQ3: How do users explain their own motivation to tweet at professional media? The theoretical background will outline potential reasons for targeting professional media using ‘@’ mentions.

The guiding premise is that whenever people look to interact with news media directly, they want to impact the content of professional news in some way. Media theory and previous

7 research presented below will explain which elements of citizen journalism can constitute a frame for the present analysis. Altogether, the answers to the three research questions would provide a clearer picture of the relations between twitterers and professional news media in the light of participatory journalism practices. With these goals in mind, the discussion is geared towards understanding the medium, the practice, the active voices in news-making via Twitter and how they challenge the agenda set by professional journalists.

3 Literature Review

3.1 Twitter: features and premises of generating news

This section introduces Twitter as a social media platform that has equally accommodated professional journalism and civic activism. Features that enable the network-wide information broadcast and news selection are discussed, with an emphasis on the types and relevance of the content generated daily on Twitter.

Since its launch in 2006, Twitter has become increasingly popular among individual users and corporate bodies, who use it as an information gathering tool, social network and branding platform (Kwak, Lee, Park, & Moon, 2010; Greer & Ferguson 2011). Twitter is an online environment where registered users can post text messages of at most 140 characters at a time. Tweets were originally designed to elicit spontaneous answers to the question “What are you doing right now?” (ibid.) Over time, microblogging has much diversified, with more original and personalized input from the users. Any line of text can now be made into a tweet; additionally, it may contain external hyperlinked content or visuals that enrich its meaning. Because it can tell and show more than just any 140 characters, a tweet can contain its message or be seen as information teaser. Only a deeper look into what individual users choose to compose or rebroadcast can tell what Twitter is made of. At a first glance, it is an information stream listing short messages which are constantly refreshed with new postings and which can be filtered based on individual preferences. Looking more closely at the content generated daily, messages posted can tell more about Twitter use in the light of current events or around influential online personas.

What is most ‘social’ about this medium is the borderless structure of the network: except for a limited number of private accounts and posts, any tweet in the network can be accessed by all registered users. With a ﬁrst boom in user numbers in 2009, Twitter reached a global audience of over 140 million active users by 2012, who in total issue over 340 million tweets per day (Twitter, 2012). A ranking of users’ popularity on Twitter is often made based on

8 their number of ‘followers’. ‘Following’ is a non-reciprocal expression of appreciation which can range from common interests to fandom: after becoming a follower, one receives in their feed all tweets of the followed accounts. To further promote a topic or endorse another user’s statement on Twitter are done by re-tweeting their post. Retweets are another measure of message popularity, which also broadens its reach and pushes it among ‘top’ posts.

The live stream of tweets can be filtered based on simple word searches, hashtagged words (#) and ‘mentioned’ (@) accounts. Simple word-based searches bring up tweets mentioning the word as such, while hashtag-based queries bring up a collection of tweets where the searched word is treated as a broader concept or topic and has a collectively recognized meaning (e.g. #nowplaying pointing out a live event on TV). ‘Trending topics’, recommended to users based on their location and interests, emerge spontaneously from hashtags with high popularity in searches and a high number of mentions. More advanced searches may bring at the forefront ‘Top’ tweets from the mass of ‘All’, following from Twitter’s algorithm of recency and number of retweets (Zubiaga et al., 2012, p. 3). External hyperlinks in shortened format and visual attachments enrich the outbound connectivity of the tweet and fully integrate it into the web. Most websites also embed Twitter or live Twitter feeds as tools for their audience to share and re-distribute content (ibid.). Judging by the type of content issued daily, Twitter has been defined as both a “social networking site and an ambient information stream” (Bruns & Burgess, 2011, p. 3). The mobile functionality of Twitter via short text messages (SMS) allows users to microblog even from offline mobile devices, which gives it more technical independence from the web and a symbolic one from potential censors. With limited space available for one-time messages, twitterers are invited to use of space and characters wisely. Technically, this constraint encourages concise messages and space-effective calls to action. Hyperlinks, most often shortened ones, can also be included in tweets, which allows broadcasting the message across various web-based platforms, enriching their meaning or re-contextualizing them (Tsui, 2008). As information-rich sources, tweets with embedded links can be understood as a “form of data sharing and as a system for creating a shared conversation” (Hermida, 2010, p. 303). Users may decode the meaning of such information teasers without following the external content. They may interpret the linked source as a conversation trigger or take it as supporting evidence for the facts summarized in the tweet.

When it comes to corporate accounts that speak on behalf of an institution, a community or a group of people, the voice that tweets is a collective one - one which is typically listened to, and ‘followed’ by many. Professional media outlets have launched Twitter accounts

9 to complement their journalistic activity in a social media environment. How and why professional media have embraced Twitter are outside the goal of this section. The way professional journalists and the medium itself motivate users to exercise their power to spread the word justiﬁes the aim of this research. Connections between traditional media moved online, their Twitter corporate accounts and their journalists’ personal accounts are evident, as these roles are connected in the original media production environment. News titles are often made into perfect call-to-action messages on Twitter, live shows gather viewers through tweeted stream links, and, conversely, the TV screens are tweeted back by viewers as attached visuals. Such normalized correlations between old, new and Twitter-media attract a growing number of followers. Currently, over a hundred media and journalists’ accounts rank among the top 1000 most followed users on Twitter (Twitter Counter, 2013), while the number of media outlets launched has grown exponentially since Twitter’s launch (Kwak et al., 2010). How twitterers source information from the feed relates to how professional media adapted the news content to Twitter. A personalized selection of news from the feed can be done by simply ‘following’ the relevant or trusted media accounts. Twitter is a already a “normalized”, convenient, way of sourcing news today (Lasorsa, Lewis & Holton, 2012) and its integration into the information routines of millions of users serves as proof.

Professional journalists or corporate media accounts use the same tools to promote their own content. Journalists may tag (@) several accounts to cite ﬁrst-hand sources and to further share their own reports (Luckie, 2012) or use Twitter as an organization tool for the newsroom (Farhi, 2009). News (re-)distribution via professional media accounts on Twitter and their followers’ reactions may generate debates around news and reporting practices. Microbloggers are a generation of media producers (Bruns, 2007) who still embrace the classic audiencing experience (Fiske, 1989) and use older media in parallel with social networks. They are able to track news stories, re-frame the viewing or reading experience and actively respond to them online.

Twitterers can share favorite news stories, contest them or debate with other users on what makes good journalism on- and oﬄine. Media commentaries on Twitter have been previously analyzed around threads with a common hashtag (Small, 2011; Page, 2012; Zappavigna, 2012), while tweeting with a critical eye on professional media has been addressed in the broader context of online news production (Burgess & Bruns, 2012). Rather than reﬂecting passively on news, users choose to contact the issuing media to voice a concern, suggest a news topic or express a critique to journalistic practices.

10 Whether it is the fascination with addressing @CNN just as texting to a friend, or just using celebrity user-tagging for self-publicity, tweeting to professional media outlets has become a growing part of individuals’ social media routines. Analyzing the news reception patterns on Twitter, Subaic and Berendt (2011) explain that a great extent of users’ reﬂection on news represents “neither creating, nor peddling [re-reporting], but extending them by commenting on news” (p. 207). This mode of news interpretation on Twitter signals a critical attitude towards professional journalism. Some of the possible explanations for media-tagged tweeting have been more recently ascribed to practices of citizen journalism (e.g. Binder, 2012; Burgess & Bruns, 2012). Tweeting to inform and to get informed are directions that develop at the same pace as the medium itself. As users develop their habits of interacting with the source of their news, professional media routines may be challenged into adopting “communalized media” (Jenkins, 2006) practices: responding to tweets as answering the much more formal letters to the editors back in the day.

Twitter is an engaging platform that easily connects audiences with professional media outlets and allows them to select, view and interact with the news content promoted in the feed. Seven years of Twitter have taught users how to make an impact on mainstream news production. The most straightforward way to contact news media outlets is by tagging them in own posts - a practice which speaks about the audiences’ expectation from their professional sources of news.

3.2 Citizen journalism

This section explores potential deﬁnitions of participatory news-making on social media platforms and relates them to the premises of the network society and the professional journalism practices. The norms and news values embraced by citizen journalists on social media are discussed, in reference to audiences expectations from professional media.

Assuming the journalist’s role without formal training is likely when audiences have a strong motivation to surpass their witnessing status and to start independently promoting information of public relevance. As “people formerly known as the audience” (Rosen, 2006) willingly join the journalistic mission, their place in the actual media landscape becomes uncertain. Potential definitions of citizen journalism range from user-generated content (henceforth UGC), which describes information supplied by untrained journalists, to actual news-making on independent platforms. The terms used by different generations of media scholars to define news-making outside professional media settings do not reconcile these nuances. With continuously changing online and offline media environments, the definitions

11 and the practice as such are extremely volatile. The challenge of defining and analyzing UGC comes from the lack of a set of norms independent of professional journalism values. Hence, whenever referring to what citizen journalism offers today, it becomes part of a normative definition, in which references to professional news values are inevitable. A more descriptive definition of citizen journalism can be done in reference to a medium of choice - in this case, Twitter. After reviewing potential directions of citizen journalism, the following sections will concentrate on tweeted news-making and the interaction between news outlets and regular users.

Wusch-Vincent and Vickery (2007) outline several criteria for producing informative UGC, including creative effort, publication, adaptation of existing content and creation outside the professional journalistic norms and routines. Less normatively, Bowman and Willis (2003) refer to the purpose, assumed audience and intentionality of citizen journalism, to define it as “an act of a citizen, or group of citizens, playing an active role in the process of collecting, reporting, analyzing and disseminating news and information.” (p. 9) Some scholars claim that a minimum of creative effort is enough to categorize UGC as participatory news-making (Bruns, 2005; Hermida & Thurman, 2007), while others view citizen journalists as reactive, rather than creative agents (Bergstr´’om, 2011). The collective aspect of news-making has received various interpretations in the literature. It ranges from “ambient journalism” (Hermida, 2010; 2011) – opinion-making around news without a specific agenda – to collaborative journalism (McIntosh, 2008) and collective intelligence (Jenkins, 2008), all put in the service of the community (Carpenter, 2008). A constant of all of these definitions surrounding citizen journalism is the environment, which should enable spontaneous news-creation and unite all voices that source and deliver news. An ideal basis for citizen journalism practice is a network structure (Gillmor, 2004) which would bring together agents that know the rules of sourcing, sharing, informing and who anticipate what is of public interest. Today social media provide the tools for distributing content and effectively link their users to exchange information in a non-hierarchical flow. Twitter meets these prerequisites - it offers a many-to-many distribution of messages and open dynamics between users worldwide.

In the light of the above deﬁnitions, aspects that stand at the basis of the network society seem to equally foster citizen journalism: “the diﬀusion of Internet, mobile communication, digital media, and a variety of tools of social software have prompted the development of horizontal networks of interactive communication that connect local and global in chosen time” (Castells, 2007, p. 246). Social media platforms that replicate the structure of the

12 network society can more directly serve the goals of citizen journalism, by allowing the quick ﬁltering and production of information, alternative sourcing and ad hoc interaction between all ‘nodes’ of the network. Twitter is one such networked environment that can also “enhance the integration of all kinds of user contributions in the making of news” (Paulussen & Ugille, 2008, p. 25), to break the classical dichotomy between news consumers and producers.

When judging by public expectations, civic duties and the social responsibility of media institutions, participatory journalism alongside trained professionals seems to pass by the norm. Although useful for the news production community, citizen-made reports are not actively encouraged by professional media, outside pre-established ‘assignments’ (Domingo et al, 2008). CNN’s iReport section, BBC’s “Have your say” and many other hosted amateur journalism sections on professional media’s sites validate users’ submissions only after professional editing. Participatory or open-source journalism (Bruns, 2007, p. 8) is usually seen by mainstream media only as a complement to their activity (Hermida & Thurman, 2008), with which they sometimes compete for masses’ attention (Goode 2009, p.1289). One of the acknowledged gains of citizen journalism in parallel, or even in competition, with professional media is “the structure of accountability it provides for traditional, ‘professional’ media.” (Reese & Dai, 2009, p. 230) This applies both in terms of broader agenda-setting process, as it will be discussed below, and spontaneously, in direct reference to isolated news stories. Whenever professional journalism is exposed in front of masses who practice citizen journalism, “fragments of information are reported, contested, denied or verified in the open” (Hermida, 2011, p. 6, emphasis added). Earlier literature has summed up under the term ‘public journalism’ (Domingo et al., 2008; Massey & Haas, 2002; Rosen, 1999) the socially responsible mission of citizen journalism, and highlighted that it aims to meet the norms of the professional media. However, tensions between different agents that practice journalism persist. Audiences have higher expectations to be “involved in the news-making process and are actively chasing discovery rather than passively being informed” (Canter, 2013), as they have access to most of the tools that professionals use. UGC, the raw material for citizen journalism, is continuously refined and better documented, so that it meets professional standards. A system of mutual accountability between citizens and journalists would resolve the question of supremacy in front of the assumed audience. It would also allow journalists and citizens to take turns at being producers and consumers while fulfilling the same mission to provide transparent news coverage.

13 3.2.1 Citizen journalism on Twitter – Premises

Citizen journalism on Twitter started with “gifted amateurs” (Lewis 2011, p. 4) who raised professional media’s awareness of topics relevant for coverage. Newsworthy topics thus supplied would then complement or fit in the official news report, and ideally, the original source would be credited. As many trending topics on Twitter are news-related (Kwak et al., 2010), a large body of research explores the ‘making of’ and the evolution of news stories in the microblogging feed. Over the seven years of Twitter, users have developed effective practices to raise awareness in front of other regular twitterers and the professional media. Users have become “social sensors of the news” (Sakaki et al, 2010) and gained the skills to expose issues new to the public eye. Twitter activity led by social activists first broke into the official coverage during the Iranian protests in 2009 (Knight, 2012) and peaked during Arab Spring (e.g. Papacharissi & de Fatima Oliveira, 2012), thus establishing several iconic moments for Twitter-led democratic empowerment (Tufekci & Wilson, 2012).

Nowadays, being ‘Twitter-savvy’ is the new information-savvy, as active users “rely on their own social filters to process news, in addition to those of traditional media (Hermida et al. 2012; Purcell et al. 2010, added emphasis). When they wish to reach beyond professional media filters, the tools available serve them well to start a new news thread. The first 140 characters of a news story may begin on Twitter, but its reach most often depends on who pays attention to it and pushes it into the official coverage. The motivation to contribute to professional coverage with a newsworthy fact relates to both individual interests and the compelling factors in the social media community:

“Users may be most compelled to contribute to news production and distribution online, not necessarily when they are portrayed by news organizations as co-creators of news, but when the way in which they are contributing is both engaging to them and smoothly embedded into their everyday lives.” (Holton et al. 2013, p.12)

The experience of news consumption and the comfort of social media environments are among the enabling factors when addressing professional media on Twitter to suggest a topic. Personal motivations may explain more about the exact newsworthy facts selected by users and how they relate to users’ attitude towards professional journalists’ selection. This is where the current study will go more in depth using twitterers’ own testimonials as sample.

When looking at what is expected to become of a tweet sent to the inﬁnite stream, the

14 assumed audience factor plays an important role. Both professional journalists and active twitterers who claim a spot in the coverage relate to the wider public’s expectations when producing a news report: “even users who do not post anything scandalous must formulate tweets and choose discussion topics based on imagined audience judgment” (Marwick & boyd, 2011, p. 124). The assumed receiver, who may or may not see a news-related tweet, helps construct a Twitter news agenda. Compared to media outlets, who may easily replicate a story headline into a tweet and publicize it live, user-made tweets are the result of a more careful selection of topics, of relevance for the producers themselves or their own assumed audience. From a technical point of view, Twitter equally accommodates topics that are pre-made in the newsroom and users’ original input. However, due to extensive reliance on external content via embedded URLs, Twitter is often seen as a “dissemination platform for legacy media” (Artwick, 2013, p. 14). Twitter-born citizen journalists may see the feed as a place where they can break into the routines of legacy media and create news stories that can later impact the newsroom culture in a broader sense. Such optimistic premises for citizen journalism on Twitter invite to a more detailed exploration of what happens in the feed, where professional media outlets are millions of followers ahead of regular users. Different forms of impacting the practice of professional journalism on Twitter are further discussed. These range from breaking media filters, or the gatekeeping process, to covering new news stories and to reconfiguring the agenda-setting process in general.

3.2.2 Citizen journalism on Twitter – Implications

This section connects existing theory on gatekeeping, news values and agenda-setting, as main coordinates of the journalistic practice, with the environment for news production oﬀered by Twitter to virtually anyone. Potential ways to impact these norms are discussed in relation to Twitter dynamics and users’ intensifying connection with the news-making culture through the feed.

3.2.3 Gatekeeping

In the classic media theory, gatekeeping defines the professional media’s practice of filtering content prior to public broadcast (McQuail, 1993: 213-14; 2005, p. 308), which affects the structure and the nature of news actually available to the masses (Shoemaker & Voe, 2009). Gatekeeping is challenged by users on Twitter who want to reach out to what is left behind the professional journalists’ filters. Active twitterers are often trained media critics, since their previous media consumption experience entitles them to recommend other news or contest existing ones. Active media criticism and commentary around existing news topics

15 are expected and often encouraged in a social network: “media criticism becomes less an organizational activity and more of a practice embedded in the citizen journalism network.” (Reese & Dai, 2009, p. 223). Critical shouts from diﬀerent ‘corners’ of Twitter can target professional news outlets directly, simply with a user tag.

Today, the countless possibilities to react to covered news suggest that a more effective and transparent way to adjust the gatekeeping process is by “publicizing rather than controlling information, under conditions of information abundance in contrast to the scarcity of the pre-digital age.” (Goode, 2009, p. 1295) Publicizing is also the default mode of sharing personal updates or news on Twitter. In a concise and spontaneous way, information takes shape under the eyes of many, without any interposed filters other than the personal interests of the ‘followers’. A more recent understanding of the gatekeeping process in media studies suggests ‘gatewatching’ as an updated definition: “an alternative to gatekeeping [] that works by harnessing the collective intelligence and knowledge of dedicated communities to filter the newsflow and to highlight and debate salient topics of importance to the community” (Bruns, 2007, p. 5). By loosening the filters of news selection already in the newsroom, professional media are likely to be more receptive in open environments, like the social media, where a forum of debate can be established ad hoc.

Gatewatching, instead of gatekeeping, may get twitterers’ hopes up when it comes to providing active critique or suggesting diﬀerent news stories. Some scholars suggest that there is still high resistance from professional journalists to let UGC pass the gates, in the light of their social responsibility norms and “due to worries about the accuracy, credibility and quality of user-generated content. (Singer, 2009, cited in Canter, 2013). If journalists’ openness to new content and receptivity to critique are higher in an informal setting like Twitter, addressing direct commentaries at(@) media can serve to at least test audiences’ awareness of news reporting tactics.

Implications of using Twitter as a tool to criticize media coverage include enhanced awareness of journalistic means and a stronger motivation to create newsworthy content that may escape professional media’s ﬁlters. In practical terms, what is expected of the tweets addressed to professional media are questions like “why not cover this topic” or “why is this news”, in an attempt to adapt the ﬁlters established in the newsroom of each media organization. When this tendency is carried further, new news topics may emerge, reproducing or adapting the news selection patterns of professional journalists.

16 3.2.4 Newsworthiness on Twitter

Active attempts to provide relevant information have already become part of recent definitions of Twitter as a social broadcasting environment. Rui and Whinston (2012) describe Twitter as “a marketplace where people contribute information to attract attention and contribute attention while consuming information.” (p. 322). More specifically, scholars conceptualize the users’ social broadcasting as a form of trade: sharing informative content to receive attention or gain popularity. In this exchange, influential Twitter accounts are targeted. Professional news media are top choice addressees and mentioned (@) accounts when it comes to broadcasting facts or opinions of potentially broader relevance. Existing news content or the high reliability on professional media’s powers inspire users to generate their own (new) news. Success stories become trending topics. Previous empirical research conducted on a large corpus of tweets categorized trending topics based on their ‘triggers’ and identified news, current events, memes, and commemoratives as main generators (Zubiaga et al., 2011). All these types have informative value and would meet newsworthiness criteria in a traditional media setting.

Different generations of media scholars and practicing journalists have refined and redefined what makes a newsworthy story. Early studies analyzing newsworthiness criteria in foreign reporting (Galtung & Ruge, 1965), domestic news (Gans, 1979) or, more generally, the construction of reality in the news (Tuchman, 1978) highlighted the selection bias towards certain communities, locations of events, either through inclusion or exclusion. The long-term significance of news stories (Gans, 2004), the novelty (Grabner, 2006) and exceptionality factors (Bouvier, 2007) and audiences’ demand for deviant and socially significant stories (Shoemaker & Cohen, 2007) are other criteria that decide what news are selected. Online news production and especially broadcasting through social media challenge these parameters. As shown before, the immediacy, transparency and abundance of information require news organizations to adjust their gatekeeping strategy and to concentrate on active users’ expectations. When virtually anyone can be a source or a news reporter on Twitter, newsworthiness is negotiated out in the open, without prior consideration of any explicit criterion. That is why an inductive approach is most suitable when discussing users’ own selection of topics. In the empirical part of this work, a classification ‘within’ will help establish several directions, rather than criteria, of newsworthiness on Twitter.

Facts considered newsworthy are published spontaneously, attached to a relevant hashtag and/or addressed to the media. What happens further is diﬃcult to track. While sometimes

17 audiences are exploited for free content (Ornebring, 2008) and the facts supplied make it into the oﬃcial news, in other cases UGC is seen as uninformed and extremist (Wahl-Jorgensen, Williams & Wardle, 2010). It is again in the power of professional media to decide what is worth covering. Ignored or not, stories promoted by citizen journalists into the feed can be seen by ‘side audiences’. If not the media representatives, people who search for media-related content in the feed can see and acknowledge these contributions, so that the twitterers’ power to raise awareness is not undermined.

3.2.5 Agenda-setting

Daily attempts to go behind the newsroom filter and original stories supplied by regular users suggest that citizen journalism on Twitter moves towards adjusting the official news agenda. Classic agenda-setting theory (McCombs & Shaw, 1972) and more recent reviews of it state that “the news media influence the salience or prominence of small number of issues that come to command public attention” (McCombs, 1997, p. 433). Moreover, the order of presentation, or relative salience, influences how important news events, issues and individuals (McQuail, 2005, p. 548) are for the masses. Nowadays, when active audiences use all available media to break the barriers, the selection and the order of reporting, setting the agenda can become a challenge for newsrooms. Citizen journalism as a civic mission is often seen as a direct response to the agenda-setting prerogatives (Antony & Thomas, 2010, p. 1281). Social networks might offer the most flexible environment to rethink the agenda. Already at the beginnings of Web 2.0, its potential to subvert the “vertical, top-down, passive, one-way flow of information” (Birdsall, 2007, p.2) was highlighted. Possibilities to renegotiate the order of presentation in the news, one of the main coordinates of agenda-setting, come with default Twitter functionality. The medium itself challenges the concept of salience in the news: it does not function as a scheduled newscast show based on intervals, but immerses all information bits available into the flow.

Optimistic views suggest that professional media already embrace, to diﬀerent extents, “new news agendas set by online and wireless citizens providing texts and images to global news stories (Wahl-Jorgensen et al., 2010 in Blaagaard, 2013, p. 2). If news organizations do not see citizen journalists’ intervention as a counterbalance to their agendas, any audience-driven changes may increase their popularity as news-makers (Tan, 1985). While news organizations compete for audiences on national markets, the globalized online environment may set diﬀerent incentives when it comes to featuring UGC in the news.

18 As follow-up to the already set agenda, news consumers are inﬂuenced to further decide what counts as useful, relevant news for themselves, a phenomenon deﬁned in the literature as second-level agenda setting (Ghanem, 1997; McCombs et al., 1998). Social media users may have the tools to adjust the news agenda; however, if they were given the chance to build a completely new agenda, they would still base it on their media consumption experiences. According to the second-level agenda setting premises, citizen journalism would not help escape the loop of professional journalism agenda. Goode (2009) demonstrated that user-generated content has reproduced the professional news selection principles, in absence of any guidelines or constraints from trained journalists.

This study looks at news selection patterns from within the Twitter feed and categorizes news genres proposed by the users themselves. Since agenda-setting is a causal theory (Scheufele, 2000), previous empirical studies surrounding this process were conducted on large samples and with predominantly quantitative approaches (e.g. Roberts, Wanta & Dzwo, 2002; Wanta & Ghanem, 2007; Boczkowski & Peer, 2011). By looking closer at individual motivations to tweet at professional media and conducting a qualitative analysis, more nuances and individual preferences will surface.

3.2.6 Some Broader Implications of Citizen Journalism on Twitter

As news values, journalistic routines, gatekeeping and agenda-setting processes are reinterpreted by Twitter users, the medium has been regarded as a tool to challenge knowledge hierarchies: “Twitter questions a news culture based on individual expert systems over knowledge-sharing” (Hermida, 2011, p. 2). It is only a matter of circumstance how the roles of news producers versus audiences are assigned. Those who share the news first and those who retweet it are equally entitled as messengers and knowledge-carriers. As a communication tool alone, Twitter already “disrupts established concepts of communication, prevailing notions of space and time and the distinction between public and private spheres” (Arceneaux & Schmitz-Weiss, 2010, p. 1265). Everyone starts from equal premises when setting up an account on Twitter; later, the influential voices and anonymous accounts share the same arena that does not require explicit ‘public’ or ‘private’ labels. The information stream allows personal nuances and provides coherent ways of filtering messages. From mere functionality to becoming a tool dedicated to citizen journalism, Twitter has taken important steps, but the opinions surrounding its potential still diverge.

19 Skeptics of citizen journalism contest the empowering potential of the Internet, and treat it as unrealistic and deceiving (Papacharissi, 2004; Wu, 2012) when it comes to actually reconfiguring power structures. Others acknowledge that civic journalism can serve as a vital complement to mainstream journalism, but not as a substitute (Reich, 2008). In a more optimistic key, other scholars highlight the democratic advantage of news-making by active citizens: “collecting information by citizens in social media allows to approach to the readers by reporting news in a broader, more democratic, and culturally more relevant way.” (Phillips et al., 2010, cited in Zubiaga et al., 2011) Twitter breeds social awareness through collective exercises in citizen journalism. The process of analyzing and producing news indicates that new “beliefs emerge [...] out of spaces of mutual awareness in which standards evolve in a reflexive process” (Reese & Dai, 2009, p. 230). Audiences may contest journalistic norms and, although not changing them immediately, they keep professional journalists accountable across online and offline spaces of news production. At this point, exercising one’s civic powers through news-making counts more than the immediate impact on the professional news agenda.

Before citizen journalism was praised in the context of social movements, reliance on routines and traditional standards had affected the diversity of available viewpoints (Carpenter, 2008). What social media like Twitter bring is a balance of perspectives between trained and untrained journalists. From this to social empowerment and democratic values there is an important step: “democracies need citizen monitors, but not everyone needs to monitor the same thing” (Graber, 2003, p. 147). Twitter encourages this structure of accountability and cultivates a sense of duty in those that use it as a media watchdog. To be understood as a citizen journalism tool, it also needs to serve the bigger civic mission and offer “alternative perspectives, context and ideological diversity to news reports, providing Internet users with the means to hear distant voices, otherwise being marginalized.” (Allan et al., 2007, p. 387). When focusing on the interaction between official media and regular users on Twitter, all these aspects will be considered to decide if the civic impulse of news-making is genuine, or just artificially constructed with multiple user tags. The supporting framework, simultaneously a working definition of citizen journalism in this context, goes along the lines of collective participation to the news making process, be that through impacting the news agenda and the journalistic practice, or just as a reflection to the mainstream media practices from an engaged audience perspective. Collective impulses on Twitter, yet not necessarily with a network-wide momentum as with trending topics, constitute a spontaneous mode of citizen journalism that reflects users’ drive to impact in some way their news sources and even become themselves sources. The overall dimension of this citizen

20 journalism mode is not analyzed in the present study. A look within this practice, however, provides an understanding of how regular citizens who are active on Twitter process news content, interpret its meaning, re-shape and complement it with new perspectives, outside any guidelines. The ‘everyday’ element primes in this analysis and its signiﬁcance for the wider citizen journalism phenomenon will surface when identifying critical perspectives on professional journalism and user-generated news topics which would diversify mainstream media’s agenda.

3.3 Self-promotion on Twitter

This section presents potential explanations for tweeting to media outlets outside the drive to criticize journalistic practice or provide original news content. Tactics for constructing an identity on social media support the premise that some of the tweets artiﬁcially targeting news media are used for self-promotion.

Not every post on Twitter that is connected to news media outlets is an act of citizen journalism. In general, tweeting varies in meaning and intention (boyd et al., 2010), and is even more diversiﬁed by the overall speed and amount of information exchanged. As any online ‘social’ activity that requires creating an account and becoming a registered user with a virtual persona, tweeting involves an “identity presentation” (boyd, 2006). This way of constructing a social persona online is intensiﬁed when users develop self-promotion tactics that involve media-tagging. This section explores the rationale by which professional media outlets become agents of promotion for anonymous users.

A personal description on Twitter needs to be more succinct than the ‘About’ section of a Facebook account. Constructing an online persona is rather a gradual process with a slow build-up: “self-presentation on Twitter takes place through ongoing tweets and conversations with others, rather than static profiles” (Marwick & boyd, 2011, p. 116). Twitter personalities are rather induced through the messages posted, than deduced from explicit self-definitions. Kaplan and Haenlein (2010) showed that a user’s identity is presented through more or less conscious disclosure of personal information like reflections, feelings, likes, and dislikes. Even before social media platforms were launched, the construction of the alter-ego in an online environment has been described as “a carefully controlled performance through which self-presentation is achieved under optimal conditions” (Papacharissi, 2002, p. 64). Since Papacharissi’s analysis of personal homepages, social networks have simultaneously both diversified and limited the individual possibilities to build truly unique profiles online. Personal updates have been analyzed on various social media (e.g. boyd, 2007; Marwick &

21 boyd, 2011) with a focus on the strategic and individualized use of the medium. Although the username, profile picture and taglines differ from one user to another, the interface and the ways to use the medium follow the same template. Twitter offers 140 characters as a convenient field to quickly transmit a message, but would not equally accommodate a detailed social and psychological profile of every user. It is the message, more than the persona, that matters on Twitter.

Interaction with news media may be grounded in self-publicity attempts. Users can take advantage of the tagging tool to promote content that is not relevant for media production, hoping to achieve higher visibility in the feed. Although Twitter offers limited space to present a personal agenda in great detail, as compared to blogs for instance, it provides indirect tools to boost popularity. Besides attaching visuals and external links, twitterers hashtag their posts and address them to influential others to maintain a visible profile. The rationale is simple: self-promoters speculate the fact that many queries are around news content and media outlets, so they tag these popular accounts just to show up in the search results and drive online traffic to own profiles or external sites.

Being an active twitterer typically means oﬀering constant updates and sharing something relevant for one’s followers or the wider network. Sharing on Twitter can be understood in many ways, since users “experiment with a wider and seemingly more varied range of collaborative creative activities” (Harrison & Barthel, 2009). In the light of such creative activities, of potential interest for larger masses, self-promotion through media is expected. However, as the selected sample will show, there are instances where self-promotion happens just by taking advantage of popular accounts, without an explicit reason for tagging media outlets. Self-promotion is not to be understood as an intrinsically good or bad Twitter practice. It is, afterall, a form of mediatization which is tied to individual explanations that this study will later explore.

Users try to break anonymity and compensate for the brevity of the message, so they promote themselves with all available Twitter tools. Kietzmann and colleagues (2011) described such tactics as “identity strategies” purposefully built for social media, while Hearn (2008) analyzed them as meta-narratives and meta-images of the self, highlighting the self-reflective aspect. Self-promotion on Twitter, artificially ‘affiliated’ with news media, falls under both of the above descriptions. Users get at “digital intimacy” (Thompson, 2008, cited in Marwick & boyd, 2011) with major news networks and treat them as agents of promotion, although it is not clear yet how the addressee is selected or if the self-promotion

22 tactic is really eﬀective. Without any aggregate measures of gained popularity, it is up to the user to decide if the meta-narratives about the self are convincing enough for the media themselves or their (misled) Twitter audiences who come across their posts.

A study of humanitarian aid campaigns in the aftermath of Haiti earthquake, which had become a trending topic, revealed that self-promotion was pre-planned and quantified on Twitter, as “users pledged to donate money with new followers” (Smith, 2010). From such noble causes to staged appearances in front of news media, tweeting for fame is expected in the present selection of cases. In an attempt “to portray both an authentic self and an interesting personality” (Marwick & boyd, 2011, p. 122), users who tweet and tag professional news outlets most probably have a well-defined personal agenda online. As “the ability to attract and command attention becomes a status symbol” (ibid, p. 127), gaining influence and popularity through official media also signals trust and reliance on media powers when looking for influence. Even if this practice violates the Twitter etiquette, and is sanctioned when the number of unsolicited user tags is too high, it deserves special consideration in the present context. Although the medium itself is exploited, the relation established between the fame-seeker and the outlet speaks for the degree of trust in the news media as authoritative, louder voices.

The reasons behind personal updates and the general presence on social media opened up the ground to understand twitterers’ individual explanations for tweeting at media outlets. Previous research has shown that these can be strategic moves, spontaneous cries for attention or just a part of social media habits. The further selection of cases and methods used will highlight to what extent tweeting at professional media is done for self-promotion.

23 4 Research design

This chapter describes the mixed-method approach and the practical implementation of the two methods. Operationalizations of concepts for content analysis and qualitative interrogation are presented sequentially.

4.1 A mixed-method approach

The compatibility between qualitative and quantitative empirical approaches has been extensively discussed in regards to conﬂicting paradigmatic associations positivism versus the Verstehen tradition and often discredited (e.g. Teddlie & Tashakkori, 2003). Other schools recommend descriptive designs in complementarity to quantitative ones, since they enable “a more accurate, comprehensive and objective representation of the object of the study” (Silverman, 2006, p. 291). In the past, triangulation was the default mode of mixed research designs (Denzin, 1970; 1976). Denzin’s own review suggests that a mix of qualitative and quantitative approaches can “secure an in-depth understanding of the phenomenon in question” (Denzin, 2012, p. 82). A more recent umbrella term for mixed-method approaches is “multigenre crystallization” (Ellingson, 2009) aims to project the object of studies through multiple lenses. Crystallization is now praised for “embodying an energizing, unruly discourse, drawing raw energy from artful science and scientiﬁc artwork.” (Denzin, 2012, p. 83). As a general approach to data collection and the selection of objects of study, the present research assembles a quantitative design together with a qualitative interpretation, which answer separate research questions. This research looks at a genre of tweets through several lenses, in RQ1 and RQ2 through the message content, while in the R3, through the issuers’ explanation. The objects of study do not overlap, yet, together they provide a more in-depth understanding of the tweeting phenomenon at large.

This integrative perspective was present in a growing body of mixed-method studies (Bryman, 2008), where “researchers employing a concurrent mixed method design do not actually integrate the quantitative and qualitative components of their mixed methods study, but [where], their findings are presented as if the two components are quite distinct” (Bryman, 2008, cited in Hesse-Biber, 2012, p. 138). With a sequential disposition of methods and results, the present study aims to explore in a systematic way the Twitter dynamics that position the users as both receivers and co-producers of news content. The main advantage of a mixed-method design for the present case is the complementary angle provided: findings of the content analysis provide a first-hand classification and interpretation of materials; these

24 are later contextualized with a set of responses of the same genre which are systematized thematically.

4.2 Content analysis

The first step of analysis concentrates on the aims stated in RQ1. Given the type and amount of material available on Twitter, content analysis can provide an effective way to describe and summarize the matter (Neuendorf 2002, p. 15). Twitter presents textual data in unique manner: the overall flow of information is made of short text messages easily countable and trackable to their source, which are interconnected based on hashtags and mentions (user-tags), but overall readable and searchable as any other textual data. Content analysis has been especially used to relate “replicable and valid inferences from texts (or other meaningful matter) to the context of their use” (Krippendorff, 2004, p. 18). Conducting content analysis on Twitter allows organizing the volatile information stream in a meaningful way and making inferences about the tweeting practice in focus.

For the type of data at hand, the unit of analysis most suitable for classification was considered a tweet. Simultaneously an eligible “unit of enumeration” (Holsti, 1969), a tweet has the advantage of an objective definition: a text message of maximum 140 characters, which may include in-text URLs and attached visuals. The population of units to which RQ1 pertains is constituted by the total number of tweets with at least two user-tags/mentions of news media accounts issued on Twitter on an ongoing basis. Tweets with only one media account tagged may represent shared content from the news media websites, which do not actually capture a user’s deliberate interaction with or reaction to the media themselves, and thus not make the object of analysis. Having a minimum of two media addressees helps define users’ explicit intention to transmit or associate a message with the tagged accounts, which goes beyond the more passive redistribution of existing news content. At this step, the object of analysis is the message contained by the tweets issued, and not the registered Twitter users that generate them. However, the two cannot be completely dissociated in the broader context of Twitter and content production on social media. At this step of the research, content analysis allows making inferences about users’ selection of topics and their relation to the news media addressees.

25 4.2.1 Sampling and data collection

The general approach to sampling from the given population is multistage purposive sampling. Purposive or relevance sampling is typically achieved in intermediary steps, by applying a set of exclusion criteria to the population until a manageable sample is generated (Krippendorff, 2004, p. 119). Twitter searches are implicitly criterion-based queries that can help retrieve tweets based on the accounts tagged, hashtags or free text. Selecting potential news media addressees is the first step in retrieving the tweets and defining the sample. News media accounts that ranked among top 1000 most followed users on Twitter (Twitter Counter, 2013) were selected, which is similar to an intensive case sampling approach. 1 Advanced Twitter queries based on a set of fourteen potential media addressees 2 served to retrieve the tweets. Any tweet tagging at least two of the selected accounts was included in a preliminary selection. Many of the popular media addressees represent the US news culture, which may create a bias in the overall analysis. However, sampling based on the above addresses does not exclude additional media accounts which may be tagged in the same unit as at least two of the fourteen selected. 3 In that sense, the sampling procedure resembles a snowball technique. The most frequently tagged media, other than those driving the sampling strategy, are presented in the Appendix, as part of the variables included in the codebook. Moreover, having the two most important news agencies in the list of addressees, The Associated Press (@ap) and Reuters (@reuters), which are primary sources for media outlets worldwide, expands the intended reach of messages and consequently, the relevance of the present selection.

Tweets were recorded and saved offline using the NodeXL plugin to MS Excel, which also allowed metadata retrieval, including separate fields for username, hashtags, URLs and attached visuals. Repeated queries were performed weekly over one month, between March 24 and April 21, 2013. The time span is large enough to capture different trending topics and avoid excessive clustering effects (Burgess & Bruns, 2012). An English-language criterion was applied at the next stage, which yielded a set of 42,291 units. The preliminary sample,

1When several accounts were aﬃliated with the same network, the most followed one was chosen, except for @cnnbrk and @cnn, where the breaking news account is the leading news provider on Twitter with more followers than the main corporate account of the network 2@cnnbrk, @cnn, @nytimes, @time, @bbc, @huﬃngtonpost, @reuters, @wsj, @foxnews, @cbsnews, @abc, @ap, @ajenglish, @forbes 3For example, the tweet “Amar Desh Editor Mahmudur Rahman Has been Arrested! Please Tweet To International Media! @AJEnglish @BBCWorld @CNN @AP @AFP @SkyNews” was retrieved based on the search query for @AJEnglish and @CNN tags. Next to these accounts, three news outlets that were not part of the initial list appeared @bbcworld, @afp and @skynews, which makes this selected unit relevant for a larger group of media addresses.

26 ordered chronologically, was systematically reduced to a convenient size, manageable for manual coding; every twentieth tweet was included in the ﬁnal sample of N=2,145 units that was considered for analysis.

4.2.2 Operationalization of variables

A broad classification of content helps to understand the purpose of this popular tweeting practice by first looking at the messages generated. An inductive approach to quantitative content analysis is preferable for the type of material at hand and the aims stated at RQ1. However, when choosing between data-driven and theory-driven operationalizations, Boyatzis (1998) recommends treating the classic inductive-deductive dichotomy as rather a continuum. In this case, observed Twitter practices and media theory can meet in the middle to help define potential content purposes. The main directions established in the literature review suggest that twitterers are likely to address media in order to bring up newsworthy topics or comment upon existing news content. In the broader social media environment, personal updates are expected to contain (self-)reflective statements and self-promotional messages that might also be addressed at professional news media.

The type of content available allows for a quick classification of units based on Twitter-specific features. These default tools operationalize the grounds on which ‘regular’ users and news media may interact on Twitter. A set of Descriptive Indicators collect information about the Date the tweet was issued, the Username that published it, whether it was retweeted or original, and the user-tags or Mentions (@) it included. Hashtags(#), Attached Visuals, embedded Links, and their external Reference were collected in the respective variables, which can also show which tools offered on Twitter are actually used to establish a dialogue with the news media. Although not a direct part of the analysis, this set of variables served to contextualize the message and understand the overall dynamics of content production and redistribution on Twitter.

The purpose of tweeting implied from the message was captured in the Content Purpose set of variables, which divided units between Newsworthy [content], Media Commentary, Self-promotion, Statement and Unclear purposes. A more in-depth classiﬁcation of units coded newsworthy was designed as a data-driven news typology that would oﬀer the answer to RQ2. To distinguish between the nuances of media commentary – criticism, appreciation and neutral comments –, a categorical variable Tone of Media Commentary was introduced. The detailed codebook is presented in Appendix I.

27 4.2.3 Pilot study

A preliminary test of the method on a similar sample, 535 tweets tagged ‘@CNN’ excluding redistributed news content, served to refine the current coding scheme. In the pilot study, more than a half of the units were coded newsworthy. Mutually-exclusive variables used to classify news genres had not captured all the dimensions of a newsworthy tweet. Even if the message was brief, the subjects and actions referred at pointed multiple implications of a news story. Instead of dividing between public affairs and non-public affairs as previously done in agenda-setting studies (e.g. Boczowski & Peer, 2011; Lim, 2010), the updated coding scheme inductively selected frequent news types – Political, Social, Crime, Conflict, Economy, Popular Culture – more of which could be present in one tweet. Detailed definitions of the variables and explicit decision rules improved the internal validity of the study before applying the same news typology to the current sample.

Newsworthy content types

This inductively developed typology is grounded in the premise that news is not an objective mirror of reality, but a sum of selective representations of the world (Schudson, 2003) – a selection which, in this case, belonged to the twitterers themselves and was conveniently fitted in the categories described below. Units coded Political referenced political figures and nation states, as well as the actions connecting these. Politicians’ public agendas or private lives and their impact on a nation were other indicators for this variable. Social newsworthy topics included references to social groups of various sizes, from individuals, local communities, peoples to the global population as a whole. Units that referenced social dynamics, events from the life of a community and more long-term developments of social agendas, humanitarian actions and social interventions were also included in this set. Units coded for Crime referenced law violations, or publicly condemned individual or group actions, from more general human rights violation, violent and environmental crimes to specific property crime cases. The Conflict variable labeled clashes and social tensions that escalated publicly - including political divergences, wars, armed conflict, protests and public movements. Newsworthy topics coded under the Economy variable included references to financial decisions in the public or private sector, public taxation, budgetary administration at the level of institutions or at the national level. Popular Culture referenced newsworthy appearances of public persons, other than politicians, media events, local happenings and cultural events. Most of the newsworthy sub-variables denominate concepts and news types with extensive

28 theoretical background and explicit guidelines for journalistic coverage. In this case, however, speciﬁc decision rules applicable to the present sample help to classify the units. News values theory will put the ﬁndings into perspective and help compare twitterers’ impromptu selection of news to the journalistic agenda.

4.2.4 Coding

Data coding was done automatically for most of the Descriptive Indicators variables 4 and manually for the Content Purpose set. Offline data collection was accessible for filtering and tracking back the original source in the feed. A second trained coder applied the scheme to a subsample of 180 units (8.39% of N) and intercoder reliability scores were calculated for each of the variables in the Content Purpose set, with Krippendorff α ranging between α=0.63 at Conflict and α=0.92 at Tone of Media Commentary - Negative. The results of the intercoder reliability tests are presented in Appendix II. Several instances of disagreement were discussed and resolved by providing additional examples and clearer decision rules.

4.3 Qualitative interrogation – Twitter replies

Following the research aim stated in RQ3, this step of the analysis explores the individual motivations to tweet at professional news media outlets. Tweets collected with NodeXL at the previous steps allowed tracking their senders in the original feed. A number of 316 users were directly addressed with a question in a tweet-format, as a direct reply to their posts. The methodological approach at this step is exploratory and somewhat experimental, given the relative recency of the social media platform and its speciﬁc type of data. The exact method draws upon non-standardized interviewing techniques and ethnographic approaches for online media, often termed ‘netnography’ (Kozinets, 2010).

The exploratory design situates the researcher within the ﬁeld of analysis, aiming to understand a genre of tweeting in its original environment. By addressing users directly, in response to their message, informing subjects are targeted directly. A context to the problem is provided for the targeted user, as their own message is the conversation trigger. An exchange of tweets in the network requires an eﬃcient use of language and conformity to the Twitter etiquette. Individual users were asked “why do you tweet at media channels?” and a note that their reply would be used for research purposes. The attempt was to start a dialogue on Twitter about, and build “rapport” with, the subjects, as recommended for

4The Reference variable in this set was coded manually to decide the nature of the URLs embedded in the tweet, whether as Professional Media, UGC or Unclear.

29 observation studies (Taylor & Bogdan, 1998, p. 48). The method as such, although not explicitly documented in the literature, has been successfully used in previous studies (e.g. Marwick & boyd, 2011). The same tactic, directed replies to Twitter users, was reviewed positively by Sibona and Walczak (2012), who emphasize the generally high response rate and the possibility to target a relevant population of subjects.

4.3.1 Operationalization

Personal motivation to speak directly to news media on Twitter requires a functional deﬁnition for the present context. Outside explanations from behavioural sciences, the motivation to tweet is here associated with the impulse or the immediate reaction to say something spontaneously on Twitter. Users’ follow-up responses would ideally capture the motivation to tweet in the original message, or more general personal reasoning regarding media tagging on Twitter. Both directions of explanation would meet the goal of the question. A ‘why’ question in direct reference to the subject’s personal actions may be intriguing in a traditional interview setting. However, the spontaneity factor plays a key role on Twitter: subjects may see the research-driven reply as a normal and spontaneous follow-up to their post and as a marker of interest from the interviewer. The ‘motivation’ investigated at this step of analysis captures both the why and the how of the practice, which can complement the ﬁndings at the previous step.

A second dimension of operationalization pertains to the informants. In relation to the news media outlets that they tagged in their postings, twitterers are audiences and potential co-producers to the news content. Seen from within Twitter, the users addressed are the only direct sources of content. In the broader media environment accommodating consumers and producers, twitterers are both. This step of analysis had the goal to go beyond their expressed message and sees them as more than audiences. As part of a fluid exchange of roles between leaders, participants, and users of content, “produsers” (Bruns, 2007) are a new type of informants that can be fitted in new scenarios of qualitative research, on social media platforms. The goal of this approach is to ensure breadth of potential explanations, without confining the space of the responses. Giving a voice to the twittterers in relation to the original response allowed every respondent to tailor his/her reply to the medium available and even expand it when they considered necessary. This experimental approach was conveniently fitted inside the medium and elicited spontaneous replies that could be studied in the original context. The approach is also consistent with the “multigenre crystallization approach” (Ellingson, 2009) specific to the new tradition of qualitative studies.

30 The phenomenon under analysis, is thus interpreted through diﬀerent lenses created by each of the respondent who explained his/her motivation.

4.3.2 Data collection

To elicit immediate replies and ensure a high response rate, users were selected randomly based on the set of tweets collected using NodeXL, at the end of week one and three of the interval. More specifically, selected users whose original (non-retweet) messages were publicized within three days before data collection days, March 30 and April 14, were addressed a direct @-reply. The structure of the reply used to collect responses included an informal greeting, the question about motivation to tweet, the disclosure and a note of appreciation, as follows:    outlets Name at  Hi Why do you tweet media channels ? 0 0  there to  networks  Much appreciated  I could use your reply in my research Thank you !  Many thanks The subject was not addressed as an interviewee in the traditional sense, but as a member of a shared social network. Replies tailored to the each users’ tweet, designed to elicit direct responses and offering a form of reward. Each collected response was ‘Favorited’, as a form of gratification, which boosts their popularity in the network. The same ‘favorite’ highlighting tool served to group the materials on a static Twitter page linked to the account used for sending the @-replies.5

The main constraint was the brevity of the message. Yet, as the research aims at using the exact means of the platform and see the ‘within’ perspective, through the users’ eyes, this approach granted a non-invasive and transparent presentation of users’ motivation to tweet. An exact phrasing of the question for all users is not recommended on Twitter, as the reply may be automatically detected as spam. Already after addressing the personalized replies, in total 376 addressed tweets, in two separate sessions at two weeks apart, the issuing account was temporarily closed for a high number of unsolicited tags. No further replies were sent

5The page can be accessed on https://twitter.com/arissme/favorites

31 after the warning, yet several replies were collected after that date, April 15. Most of the collected responses were recorded within one day from the time the question was addressed. Most of the users (97) replied with one tweet, while others posted TwitLonger posts (3) 6, or threads of up to 7 tweets by the same users. In several instances, users asked about the goal of the research conducted, before or after exposing their own motivation. A brief follow-up was sent to explain the interest in citizen journalism on Twitter. The total number of responses (a rate of 30.85% counting unique twitterers) allowed establishing some main directions of ‘motivation’ when tweeting at media outlets. The complete set of replies is presented in Appendix IV. Selected examples are presented in verbatim transcription throughout the thematic analysis. An evaluation of the means of research will help to refine the implications of these findings for the present outlook of Twitter relations. Considerations regarding the methods chosen, the objects of analysis and the research conduct in general will help align the present findings with previously discussed theory and empirical studies, and prevent too broad generalizations.

4.4 Research evaluation

This section discusses the research conduct throughout this study in terms of internal and external validity, reliability and objectivity.

The choice of methods for the present study bridges between traditional content analysis methodology and a new approach to interviewing directly in the Twitter feed. The research framework was designed to accommodate a relatively large sample size for the first step of the analysis. The population of units pertaining to the research questions was located within the medium. Hence, findings characterize existing interactions between twitterers and news media, and not the actual audience who could potentially address professional media with news-related concerns. Despite the borderless structure of the medium, distribution of the actual global population on Twitter is uneven and constantly changing (Page, 2012). Therefore, the relevance of the present findings for audiences who would like to express themselves in front of news media, but do not do so on Twitter, is limited. For the qualitative analysis, a generalization of findings was not intended either, but the approach represented an effective way to take the pulse of the news-interested users on Twitter.

6TwitLonger is an external platform linked from the tweet where users can continue their message when they exceed the 140 characters. TwitLonger was designed to help “those who talk too much for Twitter” (TwitLonger, 2013). Although it goes against the whole purpose of Twitter (ibid.), TwitLonger, as used in this instance, shows the users’ intention to tell more about their interaction with news media

32 4.4.1 Reliability

The reliability of the research conduct was supported throughout, by intercoder reliability tests and revisions of the coding scheme for the quantitative part. A reliable way to interpret the collected replies in the last step of the analysis was by offering specific examples and not claiming an exclusive, one-way meaning for any of the responses. The entire collection is available online, where the entire thread of the conversation between the researcher and respondents can be accessed. The much-debated objectivity problem in mixed-method designs can also be questioned in the present case. In the quantitative paradigm, the quest for objective truth is the driving force, while qualitative researchers are preoccupied with offering an enriching description of phenomena. As often the case, the aim of mixed-method models is to explore objects of research from different angles, which this study has also attempted. Its findings provide reliable grounds for further investigations of this Twitter phenomenon in further research.

4.4.2 Validity

At the quantitative analysis stage, the internal validity of the model was revised by coding scheme adjustments and decision rules that allow further adaptations for larger-scale studies. Variables that stored miscellaneous units had reduced frequencies after revision, which again validated the content analysis conduct. An adaption of the model that supports variation in Twitter trends, the oscillating popularity of media accounts and users’ preferences remains to be discussed in the following chapter. For the qualitative analysis, only very few replies collected from the feed suggested that the question was not clearly phrased. Outside of those, the motivation to tweet at media, as operationalized previously, was eﬀectively measured and systematically organized in prevailing themes. Hence, the internal validity of the qualitative ﬁndings is also supported.

4.4.3 Replicability

The external validity or the validity of the ﬁndings is diﬃcult to estimate, given the volatile nature of data on Twitter. The selection of potential media addressees, as discussed at the respective stage, limited the external validity of the analysis to a certain extent. The potential bias towards US-based news media followed from the selection of most popular outlets on Twitter. This criterion actually yielded where the most intense dynamics of

33 Twitter use for interaction with the media are located. Overall, this sampling strategy makes the analysis more relevant for the medium itself, which was primarily the goal of the study. The eﬀect was also largely compensated by having news agencies among the addressees, which implies an intentional global reach behind the tweets considered for analysis.

The time span selected for analysis was broad enough to avoid clustering effects around prominent news stories. No large-scale media events polarized the news agenda during the one-month interval. However, if the same analysis were replicated at any point in time, different hashtags, news stories or media criticism would surface. The external validity of the findings is affected by sampling approach, the amount of data collected from the twitterers in reply to the query and the changing habits of tweeting in general. The discussion of findings in relation to the initial aims will put into perspective the implications of tweeting at news media, with an eye to the future practices that may emerge from the feed and may impact the culture of news making.

5 Results

This section presents the ﬁndings of the content analysis and the qualitative interrogation following the three research questions.

5.1 Content Analysis – Results

Following the aims stated in RQ1, an understanding of the type content featured in the sample can be summarized in absolute numbers. Newsworthy content was featured in 572 units (26.66%) of the sample, the media commentary in 818 units (38.13%); 234 tweets (10.9%) were coded as statements, while 427 units (19.9%) expressed self-promotion material. The rest of 4.38%, 94 units, remained unclear. Tweeting to media outlets to comment on existing news content was the prevailing content purpose. The Tweet Type, original or retweet, were represented in balanced proportions, except for the Self-promotion type that had an explicit non-retweet criterion. Embedded Links were present in 19.16% of the sample, out of which 79.96% linked to professional media sources. Visuals were attached to only 2.89% of the sample, while hashtags were present in 60.79% of the cases.Only 18.24% of the links referenced UGC. The descriptive statistics for all coded variables are presented in Appendix III. A selection of the most used hashtags and most accounts addressed in the sample is presented below.

34 Table 1: Content purpose types Content Purpose n % of N Media Commentary 818 38.13 Newsworthy 572 26.66 Self-promotion 427 19.9 Statement 234 10.9 Other 94 4.38 N=2145

Table 2: Top 10 most addressed accounts Username Times Mentioned (n) % of N @cnni 326 15.19 @bbcworld 219 10.21 @guardian 183 8.53 @nbcnews 134 6.24 @barackobama 121 6.24 @amnesty 101 4.71 @ajelive 98 4.56 @afp 95 4.42 @breakingnews 78 3.63 @pbs 55 2.56 N=2145

The Tone of Media Commentary sub-variable revealed that the strongest driver for tweeting at media was providing negative criticism, with 62% of the tweets criticizing media or journalist practices. Commentaries in a neutral and positive tone had comparably low frequencies, with only 20% and respectively 11% of the tweets endorsing professional news. In line with the aims stated in RQ2, content analysis revealed the distribution of themes within the subsample of units coded Newsworthy. The set of non-mutually exclusive values allowed a flexible inductive classification of topics. The prevailing type of news referred to social causes, with 59.96% of the total newsworthy units, followed by political (37.58%), crime-related (32.69%) and conflict-related news (22.2%). Suggested news topics related to popular culture (18.53%) and the economy (10.31%) represented lower frequencies. A good fit of the data into the scheme left only 4.89% of the news unclassified.

35 Table 3: Top 10 most used hashtags

Hashtag Times Used (n) % of total hashtagged nH #bangladesh 121 9.25 #gosnell 98 7.49 #news 65 4.96 #tcot 52 3.79 #banglaspring 48 3.66 #saverohingya 40 3.05 #venezuela 38 2.91 #syria 37 2.82 #abortion 32 2.56 #viral 18 1.37

nH =1308

Table 4: Newsworthy content types

Type n % of nN Social 343 59.96 Political 215 37.58 Crime 187 32.69 Conﬂict 127 22.2 Popular Culture 106 18.53 Economy 59 10.31 Other news 28 4.89

nN =572

This distribution of content relates to the news values and agenda-setting priorities as understood by the twitterers, which will be further explored in the discussion chapter. As the focus is on the more general practice of tweeting at media, inferences will be made from the distribution of content overall, and not from the speciﬁc news topics of temporary relevance.

36 5.2 Twitter replies – Results

The following thematic analysis establishes main directions of motivation for the tweeting at professional media. These main themes were inductively derived from the materials collected. At this stage of the study, the object of analysis is represented by the responses collected from the twitterers. However, answers were elicited in direct reference to the original tweets, where multiple media outlets were tagged, which were the object of study at the previous steps. This continuity approach will permit putting the results into perspective and establishing several correspondences between the two sets of materials.

Twitterers’ responses were organized into several directions, yet, not grouped exclusively in any of the themes listed. The emerging themes presented below characterize the responses based on the actions, subjects and rhetoric used by twitterers to describe own tweeting practice.

Media critique

A set of replies emphasized the critical attitude of users towards either news media in general, or the tagged outlets in particular. Users’ main motivation for expressing a negative critique was news media’s general bias towards certain individuals, social categories or nations and the lack of objectivity in reporting: media seems to turn a blind eye to world atrocities if whites or Christians are not dying. Contrasts between various news outlets were scarce:

“I picked CNN, MSNBC, & NYT b/c they’re prominent [...] I didn’t bother with Fox”

However, such sporadic references revealed how users, from an audience position, sort news channels by popularity, quality of the journalistic act or aﬃliation with political agendas:

“those are some of the most well known #Conservative thinkers in America. (And most despised by the Left.) #tcot.”

‘Propaganda’ was a recurring media accusation that pertained to their overarching agenda, more than to speciﬁc news stories. Some already suggested solutions to the crisis of objectivity, and alternative or citizen journalism was an acclaimed practice:

“Alternative Media is light years ahead.”

37 The poor selection of news stories was a stronger reason to criticize media, than isolated journalistic practice:

“For me it’s to improve quality of content: 1) direct access to give critical feedback & 2) encourage discerning public consumption”.

More generalist statements showed twitterers’ general habit, rather than active motivation, to “keep them on their toes”: not by voicing an explicit concern about the quality of reporting, but by telling more about media’s duty and presence on social networks. Several users imagined explicit steps for how media criticism would actually reach their outlets:

“I assume they hired some intern to read everything tweeted about them (collecting data).”

Others more passively admitted that they are usually ignored:

“I know no one’s listening sometimes I like 2 yell @ the wind. #gosnell story not being covered is sick&sad.”

Twitterers also expressed an active interest in how media should translate their professional goals in front of social media audiences or co-producers:

“any responsible news/media conglomerate should be at the forefront of social media & in some way acknowledge those who address them.”

In the responses, media were referred to collectively, which might have been induced by the way the question was phrased. However, personal explanations demonstrated that users made a voluntary selection of outlets, choosing the one from which they had previously ‘audienced’ news content. The most prominent angle of media critique was unmet expectations from journalistic practice. The subjects positioned themselves as audiences with knowledge of their immediate social context and active demands for objective coverage, which had not been attained by the national and international outlets. Implicitly, their Twitter-agenda was ﬁrst to escape the status of passive news reception and make more powerful claims for objectivity from the news media. Journalists’ dismissal of potentially newsworthy stories open up a new type of motivation expressed by the twitterers, closely following a citizen journalist’s mission.

Newsworthiness

Demanding coverage for speciﬁc topics of public relevance was another motivation for contacting news media on Twitter. The approach diﬀers from user to user: while many put

38 their trust in the media to provide new coverage, others signal newsworthy topics in reference to incomplete professional news agenda. In many cases contacting media to demand coverage was part of users’ civic duty:

“I need to get the media attention in order to save #CampLiberty’s residents safety”.

In such instances, news media’s authority was not put into question; because the outlets were considered reliable or the only possible help for a social cause, they were chosen as recipients of a tweet.

It is typically social causes or conﬂict situations that elicit such demand for coverage. Speciﬁc hashtags were used to restate the purpose of the original media-tagged tweet:

“I tweet media channels in an attempt to get any response on possible coverage for #SamerIssawi.”

Presumably, success stories of Twitter as an empowering tool also inspired current respondents who subscribed to larger civic movements:

“You have to spread all movies and pictures too the world [...] My goal is to defeat Assad.”

The motivation to get mass exposure in search for justice or social empowerment also generated longer threads of responses. Subjects perceived the question already as a gratification to their initial tweet, so they were further motivated to expose the core issue that made them contact news media in the first place. More extensive news ‘folders’ opened up, such as socio-political crises in Burma, Bangladesh, Egypt and Venezuela, Catalonia’s independence, the aftermath of Benghazi attacks, human trafficking. Respondents had an active interest in the topics and their motivation to tag the media was intrinsically related to the topic in itself.

A few respondents justiﬁed their media tagging habits with the fact that they have already received attention from the news outlets themselves:

“I bring interesting stories to the attention of the other media outlets... they use some of them as ﬁll-ins.”

Twitterers are thus motivated to continue this practice, which became part of their social media routines. Although background to how the information supplied by the twitterer was included in the news report and if the source was credited is not available, the practice

39 proves to be rewarding for some microbloggers, who might see it as a daily exercise in citizen journalism. Tagging media to ask for coverage had already opened a platform for debate even before the topics suggested received attention, if any, from the media themselves. This open space and networked discussion of current events relates to another expressed motivation to tweet at media, that is to increase one’s own visibility and the reach of their message in the network.

Visibility and self-promotion

Users’ fascination with micro-celebrity on social media generates elaborate strategies for self-promotion, only indirectly connected to the professional news media. Twitterers expressed openly their intention to gain visibility or more followers without commenting on their relation to the media themselves:

“cos many people might see it (like you did)”.

They take advantage of the accounts of popular outlets, not of the institutional or journalistic component of these, to spread a message more widely. The message might not even have an assumed audience:

“tweeting to media is comparable to a message in a bottle.”

However, its reach might have some echoes in other nodes of the network. Spreading ideas, worthy or not for coverage by the tagged media, is practice for eﬀective use of Twitter’s functions.

Other respondents targeted news media to explicitly promote own platforms or news content:

I’m plannin to have my own Media Channel. It’s nt done yet, Well! It’s [JacobGreat Galaxy Media]. Broadcasting.”

Sharing the mission with reputable news media is enough of a motivation to tag more popular peers. The goal is still visibility in the network, but with a more targeted audience. The tactics for self-promotion may vary, but Twitter best practice for achieving visibility mentioned by the respondents seems to be closely related to the message they transmitted:

“Having a target or recipient gives meaning to the comment. Gets picked up in search.”

40 Such explanations conﬁrm the premises by which the study was set up – a tweet with inﬂuential addressees carries a higher order meaning and an explicit intention of the user.

Some of the voices on Twitter pointed out the quest for self-promotion of professional news media themselves:

“I believe in twitter conversations/interaction although media don’t. They are only interested in throwing news & getting followers.”

The interactivity factor is a higher order commitment demanded from the professional media. With already broad visibility, news media are criticized for their own use of social networks; thus, tagging many of them would potentially disrupt their Twitter routines and improve them with more active conversations.

Other motivations

A more general demarcation line could group responses based on how news media addressees are positioned by the twitterers themselves, whether as objects of their reference or subjects (i.e. active agents) that could carry their message further. However, the emerging themes suggested a prevalence towards seeing news media in their own right, as journalistic content producers, hence as agents, or at least promotion ‘tools’ through their accounts. In several instances discussed above the ‘regular’ user may look up – relying on their powers to cover relevant stories or appreciating their reports – or down to – criticizing existing content. It was rarely the case that the motivation to tweet should stem from a relation of equality. One user was motivated by the sharing phenomenon in more general lines, which uniquely describes a rather even balance of forces in the network.

“sharing of knowledge and information among like-minded accounts in the media industry”

Other triggers to address media on Twitter were commentaries surrounding media content, by parallel reﬂection to news and news making in general:

“adding satire to their existing headlines [...] However, I think hashtags might be working better”

Users exercise their judgement creatively in relation to the media in general, which sometimes becomes a direct object of reference in their tweets:

“Because I posted a Cartoon ABOUT the Media in the US.”

41 Coincidental tagging of multiple outlets does not fall far from the initial sampling of units for content analysis. Some of the respondents had tweeted at media outlets to emphasize their high Klout scores, a measure of Twitter accounts popularity, which recommends them as inﬂuential accounts.

Outside the prevailing themes discussed, twitterers’ ambiguous motivation (e.g just because or why not) can also highlight how spontaneously multi-tagging and retweeting happen on Twitter. Interacting with media can be part of a habit more than of a personal agenda. In several cases where the question was not clear, follow-up explanations were sent in the same conversation thread, but had a limited impact.

5.3 An integrative perspective

The results of the analysis were presented sequentially to follow the three research questions. Several correspondences between the results can help to guide the discussion and integrate the ﬁndings among previous studies of Twitter practice and citizen journalism. Results of the content analysis for RQ1 highlight the prevalence of media commentary in the sample, followed by newsworthy tweets, personal or endorsed statements and self-promotion messages. As the question was phrased to address the distribution of these genres of tweets, a quantitative reply covers the question. Inferences from these results will be discussed in reference to prevailing negative media commentary as an attempt to impact journalists’ gatekeeping and agenda-setting missions.

The answer to RQ2 derived from the content analysis of units coded newsworthy showed that stories with multiple implications were seen as relevant for coverage. Preferred stories are distributed, in descending order of frequency, among social, political, conﬂict, crime, popular culture and economy topics The new agenda potentially set by the twitterers themselves will be discussed in reference to second-order agenda setting. Criteria of newsworthiness outlined in the literature review provides the grounds of comparison between professional journalistic practice and the news value embraced by the twitterers.

Underlying motivations for tweeting at media were collected in the format of Twitter replies. Inductively observed themes provide the answer to RQ3, which showed that the main motivation to contact media on Twitter was to provide criticism, demand coverage, achieve network-wide visibility and popularity.

42 Common directions between qualitative and quantitative findings are the tendency to criticize professional news media. The incentive to create news, as reflected by both users’ motivation and the content analysed, suggests that professional news media are voices trusted by the twitterers. Two possibilities emerge from here: either twitterers consider that the existing news agenda can accommodate their topics, or they specifically bring up new topics to change the agenda and adjust the journalistic norms overall. The results open up a discussion around twitterers’ expectations from news media in general and the potential developments of this practice.

Users’ claims of visibility and popularity in the feed were another common direction established by the two sets of results. This tendency shows how Twitter usability can be redeﬁned to serve the personal agenda of each user. News media were seen as both object of reference and publicity agents for regular twitterers, which demonstrates again how spontaneously the traditional producer-receiver roles are challenged by the social networks.

6 Discussion

This chapter discusses the ﬁndings following the structure of the initial inquiries. It proceeds from the purpose of tweeting at media, to the newsworthy content types identiﬁed and to the prevailing motivations expressed by the users. The debate also suggests complementary angles that could enrich the present research, as well as ways to improve the research design.

6.1 Tweeting at news media – Purpose and dynamics

In reply to RQ1, the prevalence of media commentary over any other content purpose type suggests that twitterers are mainly driven to interact with professional news-makers by their media consumption habits. Users are first reactive to existing news content before embracing a news co-production habit, although the two modes may alternate in the social media routine of the same user. Variations around commenting versus producing newsworthy content may be analyzed longitudinally with respect to how individual users adapt their tweeting practice in relation to their news providers. The findings are consistent with Subasic and Berendt (2011), in that the dominant tendency is to interact with media by reflective commentary around existing news, and less through creation or re-distribution of new content

Tweeting and retweeting, in relatively balanced frequencies, were used to comment on and around news. Expressing a direct commentary in line with a title published by the media

43 themselves or referring to more general journalistic practice certainly reﬂects awareness of and resonance with media practice. Endorsing another user’s criticism with a retweet describes how twitterers may create alliances pro or against media. The overarching image generated is that of united audiences, who watch, read or scroll down through the same news stories, and react to them in a collective momentum. The same sense of shared meaning via Twitter had been observed before by Hermida (2010) and boyd et al. (2010), where the retweeting was operationalized as a form of conversation between users at diﬀerent ‘ends’ of the network. Twitter reinvents and strengthens the ‘weak ties’ (Granovetter, 1973) and allows audiences across the globe to create bonds based on their converging criticism of the same news media.

What the coding scheme did not capture was the long-term relation between individual users and each news outlets mentioned in their respective tweets. A distinction between whether the tagged media have been ‘followed’, thus of interest for the users in the long run, or just one-time receivers, could provide a valuable context to the message addressed. As the focus was on the content and not the user, the relations between addressees and commentators were only mapped based on a categorical variable, measuring the tone. The predominantly negative tone of commentaries suggest that twitterers take defensive position towards the content they receive in the feed. Claims against biased media reporting were also made in the qualitative dataset, suggesting that resistance to mainstream media is part of what motivates twitterers to formulate new topics.

Contesting the gatekeeping process and the ﬁlters interposed by the news agenda is grounded in the everyday role of the twitterers as media audiences. It is based on previous, even passive, experiences of consuming media that users formulate opinion towards the news coverage. Already the number of references to online professional news, via URLs, compared to the low frequencies of outbound user-generated content, indicates that professional media helps to orient the spontaneous opinion-making process on Twitter. Tweets around issues of public relevance trigger further interpretation of news content in tandem with the media themselves and other interested users. Twitter is the medium that facilitates this reaction and blurs the boundaries between production and consumption, while creating a forum for open debate. An accommodating deﬁnition for this type of environment comes close to what Rui and Whinston (2012) described as a marketplace where users give their input around or towards more popular others – in this case, professional media – in order to receive attention.

Negative commentary or appreciation of the journalists’ work are poles around which audiences gravitate when interpreting news content. With the spontaneous nature of micro-

44 blogging, the desired impact of the message posted in the feed is hard to estimate. Twitterers may simply reflect on news they have just watched on TV or read online, hoping that someone from the feed would see, resonate and retweet. Others may devise a special agenda to contest bad journalism or praise media’s accomplished mission. There is a wide range of social media routines that could explain the predominance of negative commentary. All forms of commentary have a common ground - active news consumption and the assumed audience in the feed. A group of unknown others, or ideally, the news media tagged, would credit the commentary with enough attention to gratify the creative input of the users. The mechanism that triggers such reactions and the selection of multiple news addresses are worth considering in a user-centered content analysis approach. The predominant negative critique suggests that disagreement or disappointment with news reporting correlates with the spontaneous tweet-response, better than ‘thank you’ tweets sent to the editors. That is not to say that Twitter induces negative commentary towards media, but that more control variables are required to locate the triggers of such criticism. Responses from the professional media to regular users’ tweets could make the object of a follow-up study. The effects of Twitter-mediated criticism on journalistic practice can open up a discussion regarding how social media users may impact journalists’ perceived responsibility in front of the online audiences, chiefly whether it creates more pressure in changing the agenda as compared to criticism sent via offline channels or online ones outside social media.

Newsworthy topics were the second most represented content purpose, and the type that comes closest to the active understanding of citizen journalism. Users that select news stories and bring them to the attention of professional media voluntarily undertake the mission of trained journalists, whether it is to change or to complement the existing news agenda. Twitterers tailor the story teaser to fit in a concise item that would potentially catch the eye of someone in the media. The exact nuances of the content proposed for coverage were discussed in line with RQ2. At this point, it is worth stressing that users more frequently addressed media outlets to voice a news-related concern than to promote just themselves or to make statements outside media content. To complement findings at this point, a qualitative research design may include a selection of the most active twitterers that commented on news content and/or suggested newsworthy topics, and a way of tracking their daily tweeting routines over a longer time span. This monitoring approach may be considered intrusive in a traditional field for observations. However, on Twitter it is the exact essence of a ‘following’ relationship. Mapping the long-term interactions between a selected group of twitterers and the media they follow would provide more clues as to the process of selection and appraisal of the existing news. What inspires people to look for

45 news and how a real-life story becomes newsworthy for regular twitterers are aspects that seem to be ‘educated’ daily by the audiencing habits.

As anticipated in the literature review, self-promotion tweets and social media identity strategies were present in the selection of units. Correlating media with a personal claim describes a self-interested agenda relying on the popularity of news outlets on Twitter. Self-reﬂections and meta-narratives of the self, close to what Hearn (2008) described in her study of identity strategies in the consumer culture, constituted a part of the sample that do not connect to citizen journalism per se. The trust in media, or the reliance on the personas created by the news accounts appeal to audiences who aim at fame. This type of strategy may also appear in disguise, among the tweets categorized as newsworthy or as media commentary. Newborn news channels or independent journalism projects may gravitate around the existing popular outlets to pitch their audiences when they appear in searches. Hidden tactics alike cannot be spotted based on the content-driven categorization of units. However, a closer look at the frequency of tweeting at media with the same scope may reveal more fame-seekers that market themselves on Twitter.

Personal statements that referred to media as objects of reference, outside explicit news content or journalistic practice, were present in a lower proportion to the self-promotion tactics. The collection of tweets only marginally referring to news media as ‘background noise’ to their social media routines, explains that Twitter, and more speciﬁcally, the user tagging function, can be used only coincidentally to address media. The actual @-mention is not a vocative, but a way to describe passive interactions with the news providers, most likely from an audience position. The decision rule set to categorize as self-promotion units exclusively the original tweets (non-retweets) also placed in the ‘statement’ category those messages with promotional content, but published as retweets. Showing support for others who look for fame in front of media, and retweeting them can be considered a form of solidarity in the feed, a ‘statement’ in the sense of endorsing an identity strategy. The weak ties between users across the network help to expand the reach of the messages even via unknown others, without requiring a deﬁned territory or closed circle of interaction, as it is the case on Facebook.

46 6.2 What makes news according to twitterers

The prevailing tendency among twitterers was to describe as newsworthy such happenings as from their proximal social environment or regarding social roles, groups or individuals, which would potentially impact or challenge the community they belong to. The social element prevailed in a large proportion of the sample, which confirms that the distribution of roles within the community, or even fragments of an imagined global surrounding, are often considered relevant for coverage. A challenge in interpreting this set of units comes from lacking any specific information on the actual trigger of the story: the classification does not tell more than the mere presence of social roles and individuals and does not measure in any way by what means such newsworthy stories became relevant for the twitterers themselves. Speculations can be made around the possibility to storify happenings which involve familiar social roles and individuals, by merely describing what is that they have done that is worth for coverage. Shoemaker and Cohen (2007) explained journalists’ inclination towards covering stories about the socially deviant, which comes close to what here twitterers identified as crime topics. Such newsworthy issues would expose the deviant others and serve as signal for the masses to avoid them. One of the social and crime cases of newsworthy content, distinguished with a popular hashtag, was the #gosnell story. Kermit Gosnell, the US physician, the subject of a trial on illegal abortions, whose later conviction made it into the newscast (e.g. Levs, CNN 2013), was the subject of debate among twitterers who demanded the media’s attention and coverage of the story weeks before the trial erupted in mainstream media. It was also one of the stories that re-surfaced at the qualitative analysis stage, when one of the respondents also re-attached a visual ‘proof’ of the crime story having not presented interest for professional journalists.7 The moral panics element, in the understanding of Cohen (2005/1975) is again part of what intensifies audiences’ concerns in regards to media. With the tools available today, twitterers are more likely to take position against the ‘folk devils’ (ibid.) and stir the media coverage, in order to punish and set them as outcasts. In that sense, Twitter provides the means to react and demand transparency that would create both news, in a citizen journalism sense, and a moral shield for the masses.

The deviant, negative, recent and locally proximate subjects of coverage, although the last aspect cannot be measured on Twitter, served to polarize opinions around salient

7The picture represented an empty court room: https://twitter.com/Trent Hill/status/3236899285317713 92/photo/1 to signify the lack of attention from the broader public, because professional media did not intermediate the access to the story

47 events. Even the default way hashtags are generally used is to create salience and let relevant topics emerge from the feed for those who search around those trends. Selection of newsworthy topics based on negativity was demonstrated by the relatively high frequencies of conflict-coded units. Social or cultural clashes, driven by the political background came at the forefront of the agenda configured by the users themselves. Tweets like “Police r trying 2 control the stage of. #April6 #LongMarch in #Dhaka. @Reuters @AJElive @AP @AFP @BBCWorld @cnni @ABCWorldNews @CSBNews” created clusters of meaning around real time happenings. All of the three hashtags in the above tweet configure the socio-political clash and the urge to contest the political abuses in Bangladesh. Twitter as an empowerment tool, successful in similar scenarios like the Arab Spring, inspired twitterers to even create similar hashtags, as a way to replicate the Twitter revolution to another country facing a political crisis. #banglaspring united the energy of many who sought attention from international media in large numbers – to report abuses and to support the social movement.

Negativity, as a prominent criterion of news selection, prevailed in this case, as inferred from the moderate frequencies of variables like crime and conflict. The reason for this orientation may derive from twitterers’ experience as audiences. They, like any news consumers, “flock to the media in times of crisis” (Graber, 2004, p. 563) and may become one-time citizen journalists to intensify the momentum of a potentially bigger happening. News sourced by the twitterers could be further categorized based on individuals that make the subject of news or are simultaneously used as agents for popularity. One of the popular accounts tagged alongside news media was @BarackObama, which does not exclude him being the central political subject of a newsworthy tweet. However, one of the respondents in the second part of the study suggested that tagging @barackobama ensures visibility across the network, so that one might get a chance to have their voice heard by the media if using an online persona even more influential than the media.

The degree of personalization of the news stories suggested by the twitterers could make the object of a follow-up study. As one of the news production values that guides an active issue-orientation for the public (Galtung & Ruge, 1965; Lau, 2004), the focalization of news stories on subjects with high visibility across the network may also inspire Twitter-born citizen journalists. An additional variable in the coding scheme, denominating popular subjects directly @-mentioned or implied through the message can map the degree of news story personalization. The central subjects that polarize masses’ attention, outside the information value of the tweet, may be the actual individual triggers for coverage and those who actually inspire citizen journalists in the making.

48 People’s understanding of news-making via Twitter inclines more towards ‘hard’ news (Tuchman, 1978) than towards soft topics as proven by the low frequencies of news stories in popular culture, so their shout at media is louder in moments of crisis, conflict or with immediate social relevance. Happenings that can be synthesized in a tweet to speak loud enough about the community, social roles or political challenges are preferred over sophisticated news, like those in the economy set. The findings partially confirm Goode’s (2009) skeptical view regarding the impact of citizen journalism – in that most of the topics in the ‘new agenda’ set by the twitterers replicate practices in the mainstream media. The Twitter practice as such serves as a good exercise in the making of citizen journalism; it does not, however, establish a counter-trend of agenda-setting that visibly differs from professional journalistic values. Second-order agenda setting effects, the perceived importance of news based on what the media deem newsworthy, is verified here. Explaining citizen journalistic activity on Twitter as a media effect is a pertinent theoretical framework, which can complement the present study with approaches from audience research. The temptation – or even danger – is that causal inferences based on large-scale audience research would overshadow individual nuances and reduce a twitterer’s persona to that of a passive news observer, without clear recognition of their citizen journalism effort.

This point in the discussion concentrated on the practice of newsworthy topics selection in a more general sense, accounting for time variation of topics in the news. In brief, a news agenda compiled by the twitterers would replicate many of the traditional criteria of news selection: stories with visible, immediate and negative implications for the community, often rooted in political tensions, stories oriented towards crime and conﬂict, people in power, and to a lesser extent towards the popular culture and the situation of the economy.

6.3 Personal drive to yell @ the wind

On Twitter, users’ enthusiasm, spontaneity and the permanent updates in their social media use generate routines that intersect, voluntarily or not, with news media. The established themes prevailing in the collection of replies suggest that twitterers are especially receptive to news content that does not meet their expectations of professional media. They develop tactics to change and criticize such content and assume a civic duty to correct potential media biases. The thematic analysis of replies describes the repertoire of potential roles that twitterers undertake in front of media. While many spoke of tagging media in a very narrow context reﬂecting on the tweet that triggered the conversation , others discuss it

49 as their default mission on Twitter. Users’ long-term vision for the culture of news making via social media signals that they feel empowered by the medium alone. There was no perceived impulse from professional media to keep them in a position to comment on and generate news. This conﬁrms recent observations made by Holton and colleagues (2013), showing that twitterers are more tempted to become news co-producers when they are not demanded to do so. In that sense, it can be argued that the incentive to become a citizen journalist is more powerful on Twitter in an impromptu setting, than on platforms dedicated to user-generated content. A comparison between the content generated in the feed and professionally edited citizen-news on sections like CNN iReport could reveal diﬀerences in the quality and approach to news creation.

Opposing the gatekeeping process carried out by the media becomes part of twitterers’ assumed mission. Challenging and, admittedly so, not changing, the agenda set in place by professional journalists represents an incentive strong enough for audiences to continue this practice with commitment. Demonstrating awareness of biased reporting means acknowledging the journalistic conduct and creating a counterbalance to it. What Antony and Thomas (2010) described as the main reason for embracing citizen journalism – changing agenda-setting prerogatives – is slightly remodeled on Twitter. Although they aim high with their messages, information savvy users understand that the impact of their tweet is limited and have moments when they just ‘yell @ the wind’ or hopelessly send a message in a bottle. This does not mean they abandon the mission, but rather integrate it in their social media routines.

The assumed audience factor played an important role in supporting the claim for visibility in the feed, in a similar key to Marwick and boyd’s (2011) analysis. The replies collected often referred to imaginary ‘they’ who might see, cover, acknowledge or retweet the content provided. The motivations to tweet thus correspond to the set of actions that the assumed audiences may perform in response. Twitterers were motivated to address media to be visible, because it provided a platform to strengthen their position on Twitter in a more general sense; to be critical, in order to open a space for debate and exchange opinions with others in the network; to be creative, so that they reveal what was overlooked by the media and which deserves the masses’ attention.

Personal explanations for contacting media for visibility, newsworthiness or criticism surfaced in the collection of replies from users. What the one-question interview on Twitter did not account for so far was the social desirability bias in relation to anyone, including the researcher, who gives twitterers attention in the feed. Even a question openly addressed

50 for research purposes stirred audiences’ enthusiasm, because it provided enough attention to gratify their original tweets. An improvement to this ad hoc conversation would be a longer exchange of opinions on twitter - speed-interviewing that would follow a semi-standardized protocol and looking for more complex aspects of this tweeting practice, frequency, selected addressees, media consumption habits on- and off-Twitter. Using more traditional forms of interviewing, through a conversation outside the feed, would lose the genuine aspect of interaction, spontaneity and context of conversation that proved effective in collecting relevant answers in this case. It is precisely the transparent interaction within the feed provided the context to analyze the relation between regular twitterers and the popular media accounts. The method opened up more reflective considerations around the news-making process in general and revealed which personal motivations are tactics that can be ascribed to citizen journalism.

7 Conclusions

This ﬁnal section summarizes the study and discusses its challenges and limitations. Its relevance is located within the growing literature on Twitter, while looking at potential directions for further research.

The study aimed to explore a Twitter practice that supports the premises of citizen journalism in online environments. Tagging multiple news outlets in one tweet was understood as a way by which users attempt to connect with the professional media to spread the word about stories worth covering, to comment around news, to reﬂect or to promote themselves in the feed. Citizen journalism and Twitter provided the framework of analysis. The extensive literature review established the needed connections between tweeting on a daily basis and a set of news-making standards expected of potential citizen journalists. Agenda-setting, gatekeeping and news selection prerogatives were rooted in the professional journalism norms, since deﬁnitions of citizen journalism and its respective routines vary and are also relative to the support medium.

The ﬁrst aim was to decode what is in the feed that resembles citizen journalism practices. A content analysis framework categorized the tweets addressed at most popular news outlets over one month, revealing the prevalence of negative commentary on existing news content, over newsworthy content. Tweets representing self-promotional material or statements unrelated to news content were present in even lower frequencies. Secondly, the goal was to explore

51 the nature of newsworthy tweets, using an inductive coding scheme. Tweets with a social component, those featuring political figures and the suggested coverage based on negative facts, crime and conflict news, represented the material with which twitterers would re-configure the news agenda. The news genres that surfaced in this categorization suggested that Twitter-empowered citizen journalists would compile a news selection similar to what is already in the newsroom. These everyday exercises in challenging the means of professional news media – criticism, new coverage, self-promotion tactics relying on their influence and reflective statements – showed the ‘shape’ of the imagined dialogue with media representatives, by looking at the content. Inferences based on these sets of results were made to the general practice, not the specific news preferences revealed by the current selection of cases. The interest was in the everyday dynamics and not in specific stories, which allowed extrapolations to the priorities of citizen journalism practice in general. However, a more in-depth look into the exact subjects, news stories, preferred outlets, could have revealed more of the twitterers’ everyday agenda. Follow-up designs were suggested during the first part of the discussion , including an analysis of the degree of personalization of news stories and a more explicit mapping of longterm relations between regular twitterers and ‘followed’ media.

The qualitative design aimed to reveal some of the personal motivations to tweet at media. The approach was meant to show another side of the story, not to establish a triangulation of results. Twitterers approached directly in the feed, where they had posted their message to the media, spoke about their motivation to make an impact on the news culture in general. Media critique, like in the above set of results, revealed more nuances from users’ intention to display awareness to contestation of biased journalism. The quest for visibility, often correlated to self-promotion, was part of the reason why users established a dialogue with the media. Well-defined civic missions to provide vital coverage to areas and peoples in crises were among the fervent motivations of the respondents. The thematic analysis revealed the enthusiasm of many when it comes to making an impact in front of influential media. Although the number of subjects contacted and responding was not aimed to fully represent the Twitter population, it already showed a certain convergence of views in describing the interaction with the media. People on different nodes of the network resonate in their daily exercises of citizen journalism on twitter, be those only far cries left unheard.

Conducting research on a platform that permits flexible filtering of information-rich content proved to be effective, yet challenging in terms of handling data that is volatile by its intended design. The relevance of the findings, from the distribution of frequencies

52 to the inductive themes of responses, needs to be discussed in reference to how the medium and its users evolve. The overall research design accommodated different objects of analysis – tweeted content and users’ induced replies in an attempt to create a bridge between traditional methods and more medium-tailored experimental design. The challenge was to fit in the mixed-method design separate questions that drew their answers from different, but compatible sources, and which would together provide a comprehensive picture of the practice under analysis. A leap of curiosity into what Twitter has to offer, its dynamics put to use in the form of a spontaneous qualitative interrogation could become part of the future methodology. With the right adjustments pertaining to subjects’ positioning towards the researcher, the imagined field of observations and the ways to gratify active respondents could open new ways to recruit twitterers and collect concise and targeted answers.

Methodological improvements could be made to the quantitative analysis, especially when dealing with a classification of news genre using an inductive coding scheme. A deductive approach, starting with norms of newsworthiness in professional journalism would have covered more genres and possibilities to categorize twitterers’ suggested news topics. This angle would have then allowed a direct and more specific comparison between the two agendas - the journalists’ versus the users’. The content analysis design did not discuss aspects of representation of real-life population segments on Twitter, which could be accomplished with a different selection of media addressees. The choice of media outlets tweeted at could be driven by criteria independent of the popularity factor, if aiming for a global representativity of news outlets.

Corporate accounts of media outlets may be seen as impersonal targets for twitterers who wish to cover news, complain about news or promote themselves. A follow-up research may track the interaction between regular twitterers and individual journalists aﬃliated with popular networks, when users attempt to promote newsworthy content. The degree of personalization of the tweets addressed at media representatives could be a new direction worth exploring, also in comparison to what the present research revealed.

A more in-depth discussion of what incentivizes users to address news media, and not just media in general, could have better grounded the present research. In this case, when the respondents displayed their assumed mission of citizen journalism (through criticism or new coverage), it was tied to personal experiences of audiencing. Media consumption habits can be further analyzed as triggers and then as mirrors of users’ personal Twitter-routines. Small-scale studies centered on the users’ habits can reveal if and what interest in news-making

53 is induced by the interaction with oﬄine news media.

Lessons learned while conducting this research relate first to the potential offered by Twitter and the creative capabilities of its users. Before even applying the citizen journalism filter, the ways in which active twitterers use their accounts to ‘spread the word’ proves their active critical perspective on issues of public relevance. Interacting with news media on Twitter challenges a set of dichotomies related to news production and consumption: first, between personalized reflections on public affairs and self-made news; second, between anonymous ‘use’ of journalistic material and the active discovery of newsworthy topics; third, between accepting the established news agenda and re-configuring it. These apparent opposites may blend in an environment made for dynamic interaction between all potential sources of news.

Exploring Twitter as an environment for news creation, media criticism and self-promotion helped to define new directions for understanding social media use. As a borderless network, Twitter puts in contact various users at different levels of popularity or public credibility. For the growing discipline of Twitter research this premise alone offers countless possibilities to analyze how the culture of news production is changing, if and when a clear distinction between professional and citizen journalism is required, how expectations about professional news-making are reshaped in an open environment like Twitter. Moreover, as users propagate their tweets more efficiently and assume the role of citizen journalists, their own motivation to tweet can be studied in a variety of contexts. The present interpretation revealed users’ personal agenda in relation to news media, but many more such relations between networks of users can be studied using similar approaches. The ‘what’ and ‘why’ of tweeting at news media together shape the everchanging dynamics of a captivating medium.

54 8 References

[1] Allan, S. (2007). Citizen journalism and the rise of mass self-communication: Reporting the London bombings. Global Media Journal,1(1).

[2] Antony, M. G. & Thomas, R.J. (2010).‘This is citizen journalism at its ﬁnest: YouTube and the public sphere in the Oscar Grant shooting incident. New Media & Society, 12(8), 1280-1296.

[3] Arceneaux, N. and Schmitz-Weiss, A. (2010). Seems Stupid Until You Try It: Press Coverage of Twitter 2006-9, New Media & Society, 12 (8), 1262-79.

[4] Artwick, C.G. (2013). Reporters on Twitter. Digital Journalism, Online ﬁrst, 1-17.

[5] Barkho, L. (2007). Unpacking the discursive and social links in BBC, CNN and Al-Jazeera’s Middle East reporting. Journal of Arab and Muslim Media Research, 1(1),11-29.

[6] Bergstr´’om, A. (2011). The Scope of User-generated Content: User Contributions within Online Journalism. In B. Franklin and M. Carlson (eds.).Journalists, Sources, and Credibility: New Perspectives. New York: Routledge, 167-181.

[7] Binder, A. R. (2012). Figuring out #Fukushima: An initial look at functions and content of US Twitter commentary about nuclear risk. Environmental Communication, 6(2), 268-277.

[8] Birdsall, W.F. (2007). Web 2.0 as a social movement. Webology, 4(2).

[9] Blaagaard, B. B. (2013). Shifting boundaries: Objectivity, citizen journalism and tomorrow’s journalists. Journalism, Online ﬁrst, 1-15.

[10] Boczkowski, P.J. (2004). The Processes of Adopting Multimedia and Interactivity in Three Online Newsrooms, Journal of Communication, 54(2), 197213.

[11] Boczowski, P.J. & Peer, L. (2011). The Choice Gap: The Divergent Online News Preferences of Journalists and Consumers. Journal of Communication, 61, 857-876.

[12] Boyatzis, R. E. (1998). Transforming qualitative information. Thousand Oaks etc.: Sage.

55 [13] boyd, d. (2006). Friends, friendsters, and Myspace top 8: Writing community into being on social network sites. First Monday,11(2).

[14] boyd, d. (2007). Why youth ¡3 social network sites: The role of networked publics in teenage social life. In: Buckingham D (ed.) Youth Identity and Digital Media. Cambridge MA: MIT Press, 119-42.

[15] boyd, d., Golder, S. & Lotan, G. (2010). Tweet, Tweet, Retweet: Conversational Aspects of Retweeting on Twitter. Proceedings of 43rd Hawaii International Conference on System Sciences, IEEE.

[16] Bowman, S. & Willis, C. (2003). We Media: How audiences are sharing the future of news and information. Reston, VA: The Media Center at the American Press Institute.

[17] Bouvier, G. (2007). Breaking News: the ﬁrst hours of BBC coverage of 9/11 as a media event. In: Pludowski T (ed.), How the Worlds News Media Reacted to 9/11, Spokane, WA:Marquette.

[18] Bruns, A. (2005). Gatewatching: collaborative online news production. New York: Peter Lang.

[19] Bruns, A. (2007). Produsage: Towards a Broader Framework for User-Led Content Creation. Proceedings Creativity & Cognition 6.

[20] Burgess, J. & Bruns, A. (2012). Not the Twitter election. Journalism Practice, 6(3),384-404.

[21] Canter, L. (2013). The source, the resource and the collaborator: The role of citizen journalism in local UK newspapers. Journalism Online ﬁrst, 1-19.

[22] Carpenter, S. (2008). How Online Citizen Journalism Publications and Online Newspapers Utilize the Objectivity Standard and Rely on External Sources. Journalism & Mass Communication Quarterly, 85(3), 533-50.

[23] Castells, M. (2007). Communication, power and counter-power in the network society. International Journal of Communication, 1, 238-266.

[24] Christensen, C. (2013). Wave-riding and hashtag-jumping. Information, Communication & Society, 16(5), 646-666.

56 [25] Cohen, S. (2005). Folk Devils and Moral Panic. (3rd edition, 1975). London: MacGibbon and Kee.

[26] Denzin, N. K. (1970). The research act. Chicago, IL: Aldine.

[27] Denzin, N. K. (2010). The qualitative manifesto: A call to arms. Walnut Creek, CA: Left Coast

[28] Dimitrova, D., Connolly-Ahern, C., Williams, A., Kaid, L. & Reid, A. (2003) Hyperlinking as gate-keeping: Online newspaper coverage of the execution of an American terrorist. Journalism Studies 4, 401-414.

[29] Domingo, D.,Quandt, T., Heinonen, A., Paulussen, S., Singer, J.B. & Vujnovic, M. (2008). Participatory journalism practices in the media and beyond. Journalism Practice 2(3), 326-342.

[30] Dvorak, J.C. (2006).Citizen Journalism is Like Citizen Professional Baseball: You Cant Play Pro Baseball Just Because You Think the Seattle Mariners Stink. PC Magazine, 25(18),1.

[31] Ellingson, L. L. (2009). Engaging crystallization in qualitative research. Thousand Oaks, CA: SAGE.

[32] Farhi, P. (2009). The Twitter explosion. American Journalism Review, April-May, n.n.

[33] Fiske, J. (1992). Audiencing: A Cultural Studies Approach to Watching Television, Poetics, 21, 345-59.

[34] Galtung, J. & Ruge, M. (1965). The structure of international news. Journal of Peace Research, 2(1), 64-91.

[35] Gans, H. J. (1979). Deciding Whats News: A Study of CBS Evening News, NBC Nightly News, Newsweek, and Time. New York: Pantheon Books

[36] Gans, H. J. (2004). Democracy and the News. Oxford: Oxford University Press.

[37] Ghanem, S. (1997). Filling in the Tapestry: The Second-Level of Agenda Setting. In M. McCombs, D. Shaw & Ds Weaver (Eds.)Communication and Democracy: Exploring the Intellectual Frontiers in Agenda Setting Theory. New Jersey: Lawrence Erlbaum Associates, 3-14.

[38] Gillmor, D. (2006). We the Media: Grassroots Journalism by the People for the People. Farnham: O’Reilly.

57 [39] Goode, L. (2009). Social news, citizen journalism and democracy. New Media & Society, 11, 1287-1305.

[40] Graber, D.A. (2003). The Rocky Road to New Paradigms: Modernizing the News and Citizenship Standards. Political Communication, 20(2), 145-8.

[41] Graber, D. A. (2006).Media Power in Politics. Washington: CQ Press.

[42] Granovetter, M.S. (1973). The Strength of Weak Ties, American Journal of Sociology, 78(6), 1360-80.

[43] Greer, C. F., & Ferguson, D. A. (2011). Using Twitter for promotion and branding: A content analysis of local television Twitter sites. Journal of Broadcasting & Electronic Media, 55(2), 198-214.

[44] Gunter, B., Campbell, V., Touri, M. & Gibson, R. (2009). Blogs, News and Credibility. The Association for Information Management Proceedings, 61, 185-204.

[45] Hanitzsch, T. (2004). Journalists as Peacekeeping Force? Peace journalism and mass communication theory. Journalism Studies, 5(4), 483-495.

[46] Harrison, T. M. & Barthel, B. (2009). Wielding New Media in Web 2.0: Exploring the History of Engagement With the Collaborative Construction of Media Products. New Media & Society, 11(1-2), 155-178.

[47] Hearn, A. (2008). ‘Meat, Mask, Burden’.Probing the contours of the branded ‘self’. Journal of Consumer Culture, 8(2), 197-217.

[48] Hermida, A. (2010).Twittering the News. Journalism Practice, 4(3), 297-308.

[49] Hermida, A. (2011). Mechanisms of participation: How audience options shape the conversation. In: Singer J (ed.) Participatory Journalism: Guarding Gates at Online Newspapers. Chichester: Wiley-Blackwell, 13-33.

[50] Hermida, A. & Thurman, N. (2008). A Clash of Cultures: the integration of user-generated content within professional journalistic frameworks at British newspaper websites. Journalism Practice 2, 343-356.

[51] Hermida, A., Lewis,S.C. & Zamith, R. (2012). Sourcing the Arab Spring: A Case Study of Andy Carvin’s Sources During the Tunisian and Egyptian Revolutions. Paper presented at the International Symposium on Online Journalism, Austin, TX, April.

58 [52] Holsti, O. R. (1969). Content Analysis for the Social Sciences and Humanities. Reading, MA: Addison-Wesley.

[53] Holton, A.E., Coddington, M. & Gil de Zuniga, H. (2013). Whose news? Whose values? Journalism Practice, Online ﬁrst, 1-18.

[54] Jenkins, H. (2008). Convergence Culture: where old and new media collide. New York: New York University Press.

[55] Kaplan, A. & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of social media. Business Horizons, 53 (1), 59-68.

[56] Kietzman, J & Angell, I. (2010). Panopticon revisited. Communications of the ACM, 53(6), 135-138.

[57] Kietzmann, J., Hermkens, K., McCarthy, I.P. & Silvestre, B.S. (2011). Social media? Get serious! Understanding the functional building blocks of social media. Business Horizons, 54(3), 241-251.

[58] Knight, M. (2012). Journalism as usual: The use of social media as a newsgathering tool in the coverage of the Iranian elections in 2009. Journal of Media Practice, 13, 1, 61-74.

[59] Kozinets, Robert V. (2010). Netnography: Doing Ethonographic Research Online. London: Sage.

[60] Krippendorﬀ, K. (2004). Content analysis: An introduction to its methodology (2nd ed). Thousand Oaks: Sage.

[61] Kwak, H., Lee, C., Park, H. & Moon, S. (2010). What is Twitter, a Social Network or a News Media. Proceedings of the 19th International World Wide Web Conference. Raleigh, NC, USA: ACM.

[62] Larsson, A. O. & Moe, H. (2012). Studying political microblogging: Twitter users in the 2010 Swedish election campaign. New Media & Society, 14(5), 729-747.

[63] Lau, R. W. K. (2004) Critical Realism and News Production. Media, Culture and Society, 26 (5), 693-711.

[64] Levinson, P. (2009). New new media. Boston: Allyn and Bacon.

[65] Levs, J.(May 12, 2013). Gosnell horror fuels ﬁght for abortion laws. CNN. Retrieved from http://edition.cnn.com/2013/05/12/us/abortion-trial-signiﬁcance.

59 [66] Lewis, S. (2011). The open-source ethos of journalism innovation: Between participation and professional Control. Proceedings of the 2011 Future of Journalism conference held at Cardiff University. Cardiff: Cardiff University.

[67] Lim, J. (2010). Convergence of Attention and Prominence Dimensions of Salience among Major Online Newspapers. Journal of Computer-Mediated Communication, 15, 293-313.

[68] Luckie, M. (2012). Best Practices for Journalists. Twitter Blog, September 20. http://blog.twitter.com/2012/09/best-practices-for-journalists.html.

[69] Marwick, A. E. & boyd, d. (2010). I Tweet Honestly, I Tweet Passionately: Twitter Users, Context Collapse, and the Imagined Audience. New Media & Society, 13(1), 114-184.

[70] Massey, B. L. & Haas, T. (2002). Does Making Journalism More Public Make a Diﬀerence? A critical review of evaluative research on public journalism. Journalism & Mass Communication Quarterly ,79, 559-586.

[71] McCombs, M. (1997). Building consensus: The news medias agenda-setting roles. Political Communication, 14, 433-443.

[72] McCombs, M. E., Llamas, J. P., Lopez-Excobar, E., & Rey, F. (1998). Candidate’s images in Spanish elections: Second-level agenda-setting eﬀects. Journalism & Mass Communication Quarterly,74(4),703-717.

[73] McCombs, M. (2004). Setting the agenda: The mass media and public opinion. Cambridge, UK.: Polity Press.

[74] McCombs, M., & Shaw, D. (1972). The Agenda-Setting Function of Mass Media. Public Opinion Quarterly, 3 (2), 176-187.

[75] McIntosh, S. (2008). Collaboration, Consensus, and Conﬂict. Journalism Practice, 2(2), 197-211.

[76] McQuail, D. (1993). Mass Communication Theory (3rd edition). London: Sage.

[77] ——— (2005). McQuail’s Mass communication theory. London: Sage.

[78] Murthy, D. (2011). Twitter: Microphone for the masses? New Media Culture & Society,33(5), 779-789.

60 [79] Naaman, M., Becker, H. & Gravano, L. (2011). Hip and Trendy: Characterizing Emerging Trends on Twitter, Journal of the American Society for Information Science and Technology, 62(5), 902-18.

[80] Neuendorf, K. A. (2002). The content analysis guidebook. Thousand Oaks: Sage.

[81] Ornebring, H. (2008).The consumer as producer – of what? Journalism Studies, 9(5), 771-785.

[82] Page, R. (2012). The linguistics of self-branding and micro-celebrity in Twitter: The role of hashtags. Discourse and Communication, 6(2), 181-201.

[83] Papacharissi, Z. (2002) The presentation of self in virtual life: Characteristics of personal home pages. Journalism and Mass Communication Quarterly, 79(3), 643-60.

[84] Papacharissi, Z. (2004). The virtual sphere: The Internet as public sphere. New Media & Society, 4, 9-27.

[85] Papacharissi, Z., & de Fatima Oliveira, M. (2012). Aﬀective news and networked publics: The rhythms of news storytelling on #Egypt. Journal of Communication, 62 (2), 266-282.

[86] Paulussen, S. & Ugille, P. (2008). User-generated content in the newsroom: Professional and organisational constraints on participatory journalism. Westminster Papers in Communication and Culture, 5(2), 24-41.

[87] Phillips, A., et al. (2010). Old sources: New bottles. New media, old news: Journalism and democracy in the digital age, 87-101.

[88] Purcell, K., Rainie, L., Mitchell, A., Rosenstiel, T. & Olmstead, K. (2010).Understanding the Participatory News Consumer. Project for Excellence in Journalism and the Pew Internet & American Life Project. http://www.pewinternet.org/Reports/2010/Online-News.aspx.

[89] Reese, S.D. & Dai, J. (2009).Citizen Journalism in the Global News Arena: China’s New Media Critics. In S. Allan and E. Thorsen (eds.) Citizen Journalism: Global Perspectives New York: Peter Lang, 220-231.

[90] Reich, Z. (2008). How citizens create news stories. Journalism Studies, 9(5), 739-758.

[91] Roberts, M., Wanta, W. & Dzwo, T. (2002). Agenda Setting and Issue Salience Online. Communication Research, 29(4), 452-465.

61 [92] Rosen, J. (1999). What are journalists for? New Haven, CT: Yale University Press.

[93] —— (2006). The people formerly known as the audience. Pressthink, 7.

[94] Rui, H., & Whinston, A. (2012). Information or attention? An empirical study of user contribution on Twitter. Information Systems & E-Business Management, 10(3), 309-324.

[95] Sakaki, T., Okazaki, M. & Matsuo, Y. (2010). Earthquake Shakes Twitter Users: Real-time Event Detection by Social Sensors. Proceedings of the 19th international conference on World Wide Web, ACM, New York, NY, USA, 851-60.

[96] Scheufele, D.A. (2000). Agenda-setting, priming, and framing revisited: Another look at cognitive eﬀects of political communication. Mass Communication & Society, 3(2/3), 297-316.

[97] Schudson, M. (2003). The Sociology of News, New York: W. W. Norton.

[98] Shoemaker, P. J.; Vos, T. P. (2009). Gatekeeping Theory. New York: Routledge Shoemaker, P. J., & Cohen, A. A. (2006). News around the world: Content, practitioners, and the public. New York: Routledge.

[99] Sibona, C. & Walczak, S. (2012). Purposive Sampling on Twitter: A Case Study. 45th Hawaii International Conference on System Sciences, 3510-3519.

[100] Singer, J. (2009). Quality control. Journalism Practice, 4(2), 127-142.

[101] Smith, B. G. (2010). Socially distributing public relations: Twitter, Haiti, and interactivity in social media. Public Relations Review, 36, 329-335.

[102] Small, T. A. (2011). What the hashtag? Information, Communication & Society, 14(6),872-895.

[103] Subasic, I., & Berendt, B. (2011). Peddling or creating? Investigating the role of Twitter in news reporting. Lecture Notes in Computer Science, 6611/2011, 207-213.

[104] Tan, A. S. (1985). Mass communications theories and research. New York, NY: Macmillan.

[105] Taylor, St. T. & Bogdan, R. (1998). Introduction to qualitative research methods. New York etc.: Wiley

62 [106] Thelwall, M., Buckley, K., & Paltoglou, G. (2011). Sentiment in Twitter events. Journal of the American Society for Information Science and Technology, 62(2), 406-418.

[107] Thompson, C. (2008). Brave New World of Digital Intimacy.The New York Times, 7 September.

[108] Tsui, L. (2008). The hyperlinks in newspapers and blogs. In: Turow J. and Tsui, L . (eds) The Hyperlinked Society: Questioning Connections in the Digital Age. Ann Arbor: University of Michigan Press, 70-84.

[109] TwitLonger (2013). About TwitLonger. Retrieved from May 15, 2013 from http://www.twitlonger.com/.

[110] Twitter (2012). Twitter turns six.Retrieved Februrary 7, 2013 from https://blog.twitter.com/2012/twitter-turns-six

[111] Twitter Counter (2013).Top 1000 Most Followed. Retrieved February 7, 2013 from http://twittercounter.com/pages/100.

[112] Tuchman, G. (1978). Making News: A Study in the Construction of Reality. New York: Free Press

[113] Tufekci, Z. & Wilson, C. (2012). Social Media and the Decision to Participate in Political Protest: Observations From Tahrir Square. Journal of Communication, 62, 363-379.

[114] Wahl-Jorgensen K, Williams, A. & Wardle, C. (2010). Audience views on user-generated content: Exploring the value of news from the bottom up. Northern Lights 8, 177-194.

[115] Wanta, W., & Ghanem, S. (2007). Eﬀects of agenda setting. In R. W. Preiss, B. M. Gayle, N. Burell, M. Allen, & J. Bryant (Eds.), Mass media eﬀects research: Advances through meta-analysis. Mahwah, NJ: Erlbaum.

[116] Williams A, Wardle, C. & Wahl-Jorgensen, K. (2010). Have they got news for you? Journalism Practice, 5(1), 85-99.

[117] Wu, J. C. (2012). Cultural citizenship at the Intersection of Television and New Media. Television & New Media, 20(10), 1-19.

63 [118] Wunsch-Vincent, S. & Vickery, G. (2007). Participative Web: User-generated Content. Organization for Economic Co-operation and Development Report, Paris. http:www.oecd.org/dataoecd/57/14/38393115.pdf

[119] Zappavigna, M. (2011). Ambient aﬃliation: A linguistic perspective on Twitter. New Media & Society, 13(5), 788-806.

[120] Zubiaga, A., Spina, D., Fresno, V., & Martinez, R. Classifying trending topics: a typology of conversation triggers on twitter. In Proceedings of the 20th ACM international conference on Information and knowledge management, New York, NY, USA. 2461-2464.

64 9 Appendix I

9.1 Codebook Description of variables, labels, codes, indications for coding and decision rules. A. Descriptive Indicators This set of variables describes the basic elements of the unit, in reference to Twitter- specific functions and the text content. Implicitly, these variables also capture elements of interconnectivity between users and their addressees, as well as their issue-orientation.

1. Date Date format : mm/dd/2013h:mm:ss AM/PM

The date when a tweet was originally posted; retrieved automatically

2. AssignedID Numeric variable

Sort the units in the chronologic order of their original posting Assign consecutive IDs, from 1 to 2,145 .

3. Username String variable

User who posted the tweet; retrieved automatically

4. Tweet Dummy variable type

Distinguish between originally posted by the user or retweet (marked RT) Original 0

E.g.@CNN @FoxNews @CBSNews @ABC do us all a favor and leave us alone #Rutgers

Retweet 1

E.g.RT @LUPastor: @AGChi1963 @derek˙murray93 @GMA @CBSNews @NBCNews @ABC please come watch world record broken @The˙Gridiron˙ @LeeU http://t.co/nvkSKyHQLx

5. Mentions String variable

User tags / mentions, marked ‘@’, that are included in the tweet

6. Hashtagging Dummy variable

Does the tweet include hashtags, marked ‘#’?

No 0

E.g: “@joseiswriting @nytimes @TIME As I’ve said before, lucky you. You called govt’s bluff & you won. Count your blessings. ”

Yes 1

E.g: “@Meher˙jamila @BBCWorld @SkyNews @nytimes @WSJ MQM fans, kindly use only one hashtag to trend it quickly #Vote4Patang ”

7. Hashtags# String variable

Recorded in a separate field upon retrival

E.g. #savebangladesh, #jordan, #justsaying

Dummy variable 8. Attached Visual

Does the tweet include attached pictures or embedded videos?

No 0

Only text and URLs

Yes 1

Typically marked with ‘View Media ’ option under the text of the tweet

9. Links Dummy variable

Does the tweet include any in-text URLs?

No 0

Only text (and visuals / embedded videos)

Yes 1

Includes external link(s), highlighted and underlined; continue coding for Variable 10

10. Reference Dummy variable

Professional Media 1

Links to news outlets websites or of affiliated journalists

UGC 2

Links to blogs, social media, alternative media

Unclear 3

Difficult to distinguish

B. Content Purpose

This set of variables captures the main scope and purpose of addressing the tweet to professional news outlets.

11. Newsworthy Dummy variable

Does the tweet bring up a new issue worth reporting and signals it as such OR brings up a new/unknown aspect of a theme already in the news?

No 0

E.g .“anytime the news interrupt a show I always turn to my local @ABC @wfaachannel8 or @CNN if its not on there its not #news! ”

Yes 1

E.g. “Int’l media @AJELive @cnnbrk @ABC @PressTV please be the #VOICE for #VOICELESS. Cover #March2Dhaka. ”

If ‘Yes/1 ’ continue coding with respect to the following non-mutually exclusive sub- variables (Variables 11.1 to 11.8) denominating Newsworthy Content Types

11.1 Political Dummy variable

Does the tweet include references to politicians, political issues or states as political units?

No 0

E.g. “@Ustream @CBSNews @NDTV @nbc @abchttp://t.co/iilYdDNKxz ”

Yes 1

E.g. Kader Siddique demand to arrest Our #HomeMinster for running #Genocide @AJEnglish @BBCWorld @CNN @nytimes @TheEconomist @guardian ”

11.2 Social Dummy variable

Does the tweet include references to social issues immediately impacting the life of a community, both as short-term happenings and long-term consequences – with explicit or implicit identification of social roles/groups.

No 0

E.g . “In indian politics lots of Cheamileon. @washingtonpost @WSJ @nytimes @Telegraph @BBCBreaking @TimesMagazine @FoxNews @AwazAapKi ”

Yes 1

E.g. “what about the people that cannot afford a paid repair, family’s that are struggling Xbox doesn’t care. @cnn @FoxNews @cnni ”

11.3 Conflict Dummy variable

Does the tweet include references to conflicts, wars, clashes, protests?

No 0

E.g. “Great Journalist #MahomudurRahman is in death bed.Please help. @BBC @AJEnglish @BBCBangla ”

Yes 1

E.g. “@ac360 @abc @cnn @nbc WHEN WILL YOU DO YOUR JOBS AND GO TO

#Burma??? When all of the #Rohingyaare dead?! DO YOUR JOBS!!! #SaveRohingya ”

11.4 Crime Dummy variable

Does the tweet include references to law infringement, as violent crime or property crime?

No 0

E.g. “@unow thank you for #FF and for and for being a very interesting start up. @wpjenna @jselingo @Forbes @WSJ @insidehighered #FF”

Yes 1

E.g. “Protection of human rights! #March2Dhaka #6April #SaveBangladesh @TheEconomist @AJEnglish @BBC @CNN @nytimes @guardian @RT˙com @savebdesh ”

11.5 Economy Dummy variable

Does the tweet include references to economy and finances, taxation, money-issues?

No 0

E.g. “We demand relief of Mahmudur Raman of BD right now.#savebangladesh @AP @BBCWorld @CNN @AJEnglish #PoliceState ”

Yes 1

E.g. “#Compton Ca..Corruption, incompetence or Business as usual? http://t.co/42BzmRIMQ1” @latimes @wsj @nytimes @billmaher @mmflint @robbieconal ”

11.6 Popular culture Dummy variable

Does the tweet include references to entertainment sources and products, sports events, celebrations, celebrities?

No 0

E.g. “Ask the Supreme Court why: ...a man sentenced to death/life in prison has a right to marriage but not a taxpaying Gay American? @time @cnn ”

Yes 1

E.g. “#JustinMeetMurelle @KingJames @espn @BarackObama @CBSNews @FoxNews @justinbieber @Oprah @Drakee˙YMCMB @selenagomez ”

11.7. Other news Dummy variable

All other units coded newsworthy that did not fall under any of the variables 11.1- 11.6

12. Media Commentary Dummy variable

Does the tweet represent a commentary to media content , the journalistic approach or journalists?

No 0

E.g: “David 1272. Dawn 473. Nearing 4 hours #guinnessworldrecords @cnn @gwrnews ”

Yes 1

E.g: “Watching a great round table discussion on @CNN with Senators -Rep & Dem- leaving/retiring at end of yr #wpsd http://t.co/BuN2dKeD ”

If Yes/1 continue coding for ‘Tone of Media commentary/12.1.

12.1 Tone of Numeric, mutually-exclusive values Commentary

Negative 0

E.g. @nytimes @abc @cbs Obama: “Shame On Us ” if We Forget Newtown … But Go Ahead and Forget Benghazi http://t.co/NmY083yzCA

Neutral 1

E.g. The @AP Dropped the I-word but @nytimes hasn’t. Tell the #NYT 2 Do the Same Take action @ http://t.co/M9sKDNYGBq #droptheiword #p2c #cir2013 ”

Positive 2

E.g. appreciate @NYTimes and the @AP for publicizing the Foley family’s plea to #James˙Foley’s unknown captors in #Syria http://t.co/ga1KZveJDi

Unclear 3

E.g. What you tweet matters more than how much. @NYTimes posts 2X as much as @CNN but they remain tied for followers. http://t.co/w7QMd7HXGn

13. Self Promotion Dummy variable

Does the tweet represent a personal statement, reflection or opinion with reference outside the media content?Obs: only original, non-retweets

No 0

E:g “@mainstreamedia Burn &; bury Ibrahim Ali & BN as BN supported Perkasa Ibrahim Ali by not taking any actions @fmtoday @MsiaChronicle @BBC @CNN ”

Yes 1

E.g .“In #Guadeloupe, with 27 degrees... Wassup for you guys? Time to update my blog http://t.co/M01R5moINg. @SABCNewsOnline @BBCNews @CNN @ABC ”

14. Statement Dummy variable

Does the tweet represent a personal reflection or opinion with indirect referenceto news media content?

No 0

E.g. “Who has named CrashPlan a top backup solution? @USATODAY, @NYTimes

and @WSJ to name a few: http://t.co/bScTmHdYmu #WorldBackupDay ”

Yes 1

E.g. “The disheartening truth about the arguments against same sex marriage is how ignorant people are to the actual issue @msnbc @FoxNews @CNN ”

15. Unclear Dummy variable

Include the tweet if it is not clear/ if it is none of the other ‘Content Purpose ’ Types

No 0

E.g. “RT: @CNN @BBCWorld @AlArabiya˙Eng @HuffPostWorld @AP @AJEnglish which one of you will report on @samerissawi1’s plight? #IssawiResistance250 ”

Yes 1

E.g. “Top 10 brands on #Twitter by # of followers: @YouTube @instagram @twitter @twitter˙es @cnnbrk @UberSoc @twitpic @FCBarcelona @nytimes @cnn ”

9.2 Decision Rules A. Descriptive Indicators 1. The classification Reference in the links should be based on the column from NodeXL listing ‘domains in tweets ’. For bit.ly and other URLs in shortened format open link in new tab, check content and then categorize. B. Content Purpose: 1.a When a message points at media being ignorant of a topic which the user later describes, assign the unit to Newsworthy; the negative criticism is implied, however, when the twitter himself/herself provides own description of the content for coverage, it will be considered as ‘newsworthy ’ input . 1.b. Conversely, when commentary around media topics are limited to an aspect of a story that has been already covered, it will be considered first ‘Media commentary ’, then categorized as ‘negative ’ for Tone (12.1) 1.c. For crime, code all perception of deviance/outside rule of law and if the twitterer considers the facts as ‘moral crimes ’. No need to consult state specific-laws. 1.d For conflict, code clashes and tension of any kind(cultural, political, protests, fights). Hashtags around conflict areas, such as #syria, #banglaspring can be helpful when deciding. Also include more general newsworthy messages that refer to abstract notions surrounding conflict (e.g. using #war, #stopviolence), or mentioned subjects involved in it (e.g. @amnesty). 2. When tweets include commentaries around a news title or sounding alike, next to an URL, check the external Reference before deciding whether it is new newsworthy topic or just a comment from the user (capitalization of words in the title is an easy way to spot titles from professional media, making it easy to categorize the tweet as commentary; always doublecheck). 2.a Neutral tone media commentary also includes ‘mixed feelings ’ or conflicting evaluations of the media tagged in the tweet 3. Self-promotion tweets are exclusively original tweets – re-tweeted promotional material is considered endorsed statement, and thus will be coded as ‘Statement ’ 4. When hashtags appear next to other news outlets, in addition to those @-mentioned (e.g. #cnni, #bbcworld) it can be misleading way of using them; deduce from the message if the media were intended as addressees or ar just the ‘object ’ of a personal observation. (e.g. “I watch #cnni ” would be a 14. Statement , whereas “Hey #cnni , cover this... ” could be a potential newsworthy topic)

10 Appendix II

10. 1 Intercoder Reliability Tests

Variable Krippendorff ’s Alpha (ɑ) 11. Newsworthy* 0.89 11.1 Political 0.89 11.2 Social 0.81 11.3 Conflict 0.63/0.86** 11.4 Crime 0.83 11.5 Economy 0.91 11.6 Popular Culture 0.9 11.7 Other News 0.7/0.82 12. Media Commentary 0.91 12.1.Tone - Neutral 0.81 Positive 0.91 Negative 0.92 Unclear 0.8 13. Self-Promotion 0.8 14. Statement 0.76/0.89 15. Unclear 0.8

*The coding for intercoder reliability tests purposes was carried out starting with the supra-ordinate variables (11.Newsworthy, 12. Media commentary, 13. Self-promotion, 14. Statement , 15 Unclear). After agreement scores were computed for these variables and some examples discussed, the intercoder reliability test continued with the sub-variables (11.1 to 11.7 and values for 12.1 ). **For variables with initial ɑ<0.8, coding instructions were revised and more explicit decision rules developed. A set of 30 newsworthy units were coded for 11.1-11.7 in a second reliability test.

11 Appendix III - Descriptive Statistics 11. 1 Descriptive Indicators

Table 1: Distribution of Content Purpose over Tweet Type Media Newsworthy Self- Statement Unclear n %of N Commentary Promotion* Tweet 478 112 427 202 36 1256 58.55 Retweet 340 459 0 32 58 889 41.44

*See codebook and decision rules : only original tweets were coded as self-promotion

Table 2: @-Mentions – Top 10 most addressed user accounts

Username Times %of N Mentioned(n)

@cnni 326 15.19 @bbcworld 219 10.21 @guardian 183 8.53 @nbcnews 134 6.24 @barackobama 121 5.64 @amnesty 101 4.71 @ajelive 98 4.56 @afp 95 4.42 @breakingnews 78 3.63 @pbs 55 2.56 N=2145

Table 44: Frequency of Hashtagging

Hashtagged n % of N Yes 1308 60.97 No 837 39.02 N=2145

Table 55: Top 10 most used hashtags

Hashtag Times %of total

Used(n) hashtagged n H

#bangladesh 121 9.25

#gosnell 98 7.49

#news 65 4.96

#tcot 52 3.79

#banglaspring 48 3.66

#saverohingya 40 3.05

#venezuela 38 2.91

#syria 37 2.82

#abortion 32 2.44

#viral 18 1.37 nH=1308

Table 3: Presence of Attached Visuals

Attached Visuals n % of N Yes 62 2.89 No 2083 97.11 N=2145

Table 44: Presence of Links Links n % of N Yes 411 19.16 No 1734 80.83 N=2145

Table 55: Type of content referenced in Links

Linked content n % of ni Professional Media 327 79.56 UGC 75 18.24 Unclear 8 1.94

ni =411

12 Content Purpose Table 66: Content purpose types Content Purpose n %of N Newsworthy 572 26.66 Media Commentary 818 38.13 Self-Promotion 427 19.9 Personal Statement 234 10.9 Other 94 4.38 N=2145

Table 7: Tone of Media Commentary

Tone n % of nT Positive 93 11.36 Neutral 161 19.68 Negative 507 61.98 Unclear 57 6.96 nT=818

7% 11% Positive Neutral Negative 20% Unclear 62%

Figure 1: Tone of Media Commentary

Table 8: Newsworthy topics types

Type n %of nN Political 215 37.58 Social 343 59.96 Conflict 127 22.2 Crime 187 32.69 Economy 59 10.31 Popular Culture 106 18.53 Other News 28 4.89 nN=572

77 12 Appendix IV

12.1 Collected Twitter replies

Collected replies are listed in reversed chronological order, except for the threads posted by the same user.

Commentario Politica (@CrveniTalas), 3 May I am just learning the ropes. My goal is to bring people to the website http://commentariopolitica. com I want to increase followers.

Ahsanul karim Forhad (@akforhad), 1 May I want to tell you that most of the media in Bangladesh not telecast the truth news. so I must to tweet to international media. julien dl (@juliendl), 23 Apr Truth

IMPACT YOUR WORLD (@IYWLeadership), 17 Apr we are engaged in world changing activities;tweeting@ media outlets pushes our noble ideas beyond our immediate geographical limits

Morgan Darenger (@MorgzDarenger3), 17 Apr Sorry didn’t see this till now! :P What research?

NYFunnySongs @NYFunnySongs, 16 Apr They supported us, hoped they’d RT

Sasha Kinney (@sashaki), 15 Apr For me it’s to improve quality of content: 1) direct access to give critical feedback & 2) encourage discerning public consumption

Mike (@ToSmart4OwnGood), 15 Apr to share knowledge, and true unbiased, uncensored news that our media ( Massive Ethnic Destruction In America) won’t.

Larry Miller (@LarryMiller2012), 15 Apr

78 Hej! News tip;80,000dead in Syria..8 times that if chemical war.Not only tweet also e-mail,&/or govt.Japan NUKEcrisis:suggested sand(1) sand to encase(as in Chernobyl);turnedTOglass encasing NUKE particles. Japan WATER spreadNUKEeverywhere(2)

Antonella Napoli (@AntonellaNapoli), 15 Apr of course. I’d want that US media support stronger the reform of 2th amendment. For this mom and others. Thanks for your attention.

IamEvilHomer (@Trent Hill), 15 Apr twitter caused this response to #gosnell http://www.breitbart.com/Big-Journalism/2013/ 04/13/Media-Acknowledge-Blowing-Gosnell-Story-Pledge-Extensive-Coverage(1) I know no one’s listening sometimes I like 2 yell @ the wind. #gosnell story not being covered is sick&sad pic.twitter.com/NuJOQAwndm(2)

Abbas Hamideh (@Resistance48), 15 Apr I tweet media channels in an attempt to get any response on possible coverage for #SamerIssawi. It is evidence of informing them.

Felipe Sahagun (@sahagunfelipe), 15 Apr Simple. I tweet what I consider useful for internationalists and/or journalists who share my interests (1) If I ﬁnd a documentary, article, report, book, ﬁlm... good to share with my students and colleages, I don’t care about source (2) As you can see in my tweets, I’m much more interested in learning and help others to learn than in teaching (3) Nevertheless, I try to answer any question, no matter what I think about it out of respect for any tweeter who care to ask (4) Finally, I’m convinced we are pioneers in a new communication paradigm and would be irresponsible not to be part of it. (and 5)

Johnny AppleBong (@DustinFebruary), 15 Apr Social media is where anyone can have a voice, so I can call out an account and let others be aware of the modern medias tricks

Word Monger (@Tazatator), 15 Apr

79 The contested #NDAA ’law’ that makes propaganda legal through main stream media outlets and their refusal to report objectively.

Tyler Scott( @tylerscott89), 15 Apr any responsible news/media conglomerate should be at the forefront of social media & in some way acknowledge those who address them if not personally, at least acknowledge a mass call to action, ie. regarding #gosnell

Adesina Jacob Olaolu (@AdesinaJacobsen), 15 Apr Hello ! I’m plannin to have my own Media Channel. It’s nt done yet, Well! It’s [JacobGreat Galaxy Media]. Broadcasting

Antonio Yuncal (@AntonioYuncal), 14 Apr I believe in twitter conversations/interaction although media don’t. They are only interested in throwing news & getting followers.

John Williams (@John MWilliams), 14 Apr great question! REACH I believe we can change by doing whats uncomfortable #inﬂuence is progandana so why not go to the root @CNN

Cleophis FlyJuice (II @CFlyJuiceII), 14 Apr Because I posted a Cartoon ABOUT the Media in the US.

Tess Aluise (@TessAluise), 14 Apr to get their attention and let them know that we are aware they are blocking news that does not ﬁt into their biased agenda

Voris Cunningham (@V Cunning), 14 Apr I tweet @ media channels when they fail to cover stories which are of vital importance to the public.

Istiaque Al Mahmud (@iamahmud), 5 Apr i found some international media made wrong interpretation of shahbag movement on some reports.

Egypt for All (@G2KG), 2 Apr

80 @potterarchy kidnapping of a child while leaving church in Egypt under MB government. Help bring him back. pic.twitter.com/qWPvvTKo54

Kelly (@potterarchy), 1 Apr Just helping my followers in case they’re interested in checking out one of those accounts. What research are you doing?

Peymaneh Shaﬁ (@peymaneh123), 1 Apr hi Arissme I need to get the media attention in order to save #CampLiberty’s residents safety.

Scrabby doo (@Scoobydoo007), 1 Apr You have to spread all movies and pictures too the world. By agencies you get more viewers. My goal is to defeat Assad.

shrabon dhara (@muktirgaan), 1 Apr bcoz media should know what is the actual demad of gen. ppl.

Mr ibreess (@carragha), 1 Apr just

PAPARAZZINC (@PAPARAZZINC), 1 Apr To shake them up.. give a diﬀerent point-of-view.. to upset them, show American Media not every American is gullible. . Prob?ems (@AzeriProbs), 1 Apr I usually don’t, when I do it’s because of a major event, like a genocide, for example, and I would like those medias to mention it

Shirley James (@Shirlz007), 1 Apr Also try get hold of @Pat Dickens he is heading the campaign. I think its brilliant and way overdue.

Pat Dickens (@Pat Dickens), 1 Apr @Shirlz007 Thanks Shirley. Yup any questions feel free to ask.

Shirley James( @Shirlz007), 1 Apr

81 Their website is currently down but they have launched a back-up here http://tson.weebly.com/

Shirley James( @Shirlz007), 1 Apr Hi read @TheSealsOfNam declaration of war on #sealhunt http://greenwayup.com/blogs/590/ 318/declaration-of-war-on-namibia-se or email [email protected] so can give expln

Andrew J Banks (@supprosetry), 1 Apr A sharing of knowledge and information among like-minded accounts in the media industry.

Alan Hooper (@AlanRHooper), 1 Apr My purpose here was to point out these accounts that have achieved the extremely high Klout score of 99.

Mark Anthony DiBello (@MarkDiBello), 1 Apr 1. Preach the Good News. 2. The broadcast incites a response. 3. Mass communication. Hope that helps. Peace.

Le Roux Nel (@LeRouxNel), 1 Apr Allows me to be involved in different causes & to support the prevention of democracy deficits, weak institutions & poor leadership petra (@1954candanga), 31 Mar You are most welcome! to use whichever of my tweets as long as you write down the origin or origins(1) I am tweeting to media channels to spread whats happening in my country, there are many abroud who write about late Pres #Chavez ,(2) As if he was a ”good” President. The real scene is he became very very rich, he used poverty as a tool to dominate the masses,(3) He committed high treason,he allowed cuban flags all over the country, at the moment Cuba s natl anthem is being sung on state TV(4) He ruined the country, at the moment supermarket shelves are mostly empty like in #Cuba This has NEVER happened, we have underwent(5) two devaluations of our currency, the bolivar, in 50 days, a very rich oil country has been ruined!(6) So, to avoid wrong ideas about my country overseas, I tweet to the media. Kind Regards from #Venezuela(7)

82 Melody Stuart (@MStuart1970), 31 Mar Not sure what you’re asking. I want media 2 stop pushing progressive agenda & report the news fairly. What LSM omit is telling. alasscan (@ak2sandiego), 31 Mar They monitor twitter, they know what is trending, most are active on twitter. The pipeline rupture isn’t getting much coverage.

Chad Zahner (@Chad Zahner), 31 Mar to get them to report news, I love news and hearing it twitter gives me the ability to (cont) http://tl.gd/lervit [ read it because honestly mainstream never reports it or they are hours or days behind] (1) truth be told I am motivated for. Better news on all media (2) enric sanchez ——*—— (@duxans), 31 Mar I tweet the media in order to publicize the Spain’s dirty game against independence for Catalonia

Jeﬀ Friedman (@PhillyPhriedman), 31 Mar calling out media outﬁts is a way of getting their attention :)

Yervant Terzian (@yterzian), 31 Mar Hi Diana. I just happen to be interested in genetics and more speciﬁcally rare diseases. Cheers.

AM:Hate Radio Inc.( @AMHateRadio), 31 Mar Expose #Truth That #Democrats & #Republicans Are Controlled By #CorporateNewsMedia #Plutocrats & There’s No Liberal Media 100% Hoax (1) Late George Carlin Re:#CorporateNewsMedia ”They Own All The News Media Outlets -They Got You By The Balls” http://www.youtube.com/watch?v=acLW1vFO-2Q&feature =share (2)

ALBAIH (@khalidalbaih), 31 Mar they didn’t reply! What’s your research on? (1) lots I ideas :)(2)

83 Discover Somaliland (@DiscSomaliland), 31 Mar What are you researching on?(1) Hello. We are in the process of putting a plan together for 18th of May; contacting as many media channels as possible.(2) The likes of #skynews #mbc #HCTV #BBC had previously covered May 18th #Somaliland independence day.(3)

kingsGeorge King’ori (@kings ori), 31 Mar hi,cos many people might see it (like you did)

Rashid Aurakzai (@RKzoy), 31 Mar To tell’em that there r more important, real & deserving issues 2 cover than what they’re broadcasting. U can use.

Harri V (@uAlrdyKnoWaItIz), 31 Mar I am adding satire to their existing headlines..so targeting similar audience. However, I think hashtags might be working better?

Nader Moaveni (@nmoaven), i31 Mar media need to hear from us in order to set their priorities better in reporting the important issues so people in power get it R8!

Faree (@Kalhukoe), 31 Mar just to get there attention .. Most of the time they r ill informed ... Hope im doing the right thing ... Tks

Egypt for All (@G2KG), 31 Mar to bring their attention to facts they may not be aware of. They may be able to help in exposing dictator through their channels

Meher Jamila (@Meher jamila), 31 Mar Motivatn To TAG Media Chnls 2 Inform #World Real Current Situatn Inside A Vurnerable ATOMIC POWER& Concequences OF Rapidly 1/2

Osama Obama (@IranAware), 31 Mar

84 . keeps them on their toes..

Kushan Buwa (@kbuwa), 31 Mar because most of the media channels are biased.. I want them to understand where they stand.. 140 isn’t enough

Wesley Sargent (@TaxesYouPay), 31 Mar Sure! You’ll see I have 3 tweets daily regarding #Benghazi. I started w/ 1, no media mention, just drawing anyone’s attention to...(1) to # of days that had passed since event. Then one day I started seeing similarities to Watergate. Then I added the 2nd tweet. (2) Sometimes I tweet them all & say, ”First 1 to interview a #Benghazi Survivor gets the Pulitzer. GO....” Call it ”baiting the hook”!(3) Once I added Watergate, I thought it was a natural progression to tweet the media, looking for a 21st century Woodward or (4) or Bernstein. Thought it would be interesting to tweet the media & see what happens.(5)

Abdul Alim Masud( @human ﬁgure), 31 Mar thanks

Amanda Rose( @TheAmandaRose), 31 Mar If it is relevant to my followers I retweet or tag. If I value their feedback and opinion I will also tag and mention.

UnoﬃcialTranslator (@lcc19blueridge), 31 Mar It is a random reply tweet of ”twitter Bot”. Please allow by the tolerant gentleness .

Lenin (@nine L), 31 Mar we are looking for world media coverage of our movement. Thanks. /cc @Afroza13 @sumonsm00 #shahbag #RumiSquad

Nicola Tenerelli (@NicolaTenerell)i, 31 Mar Hi Diana, I want tooﬀer a diﬀerent point of view by italian massmedias

Pottles (@MyPottles), 31 Mar Media attention speaks volumes on current events & social issues that concern prohibition

85 + it paves the road for open discussion

V for Vendetta (@Texgalleslie), 30 Mar Our country is a Republic and not a Communist country like our Media thinks. What is reported is what Obama wants. Little truth.

aqv (@aqv21), 30 Mar A single verbal confrontation, using truth as a weapon, is like one bullet ﬁred at the enemy in war. - Kyle-Anne Shiver

BD Thinker (@bdthinkers), 30 Mar thnx. why not? may i know u and abt ur research?(1) actually my friend, I just only learn twitter within not more than 1 month though i’ve signed up (cont) http://tl.gd/n 1rjgbhb [two years be4, but it attracts me much after i’ve learn abt it. btw my experience says, media response tweets when they get news with reference.](2)

Mechatronic Shudipto (@nBUET), 30 Mar Hi,Diana. I tag Intl media 2 let them know what’s happening in my country.Also share things which local biased media don’t publish

¡illegible string¿(@Dasein only), 30 Mar The Story is worth millions, you are welcome to bid? Rupert Murdoch is in the frame.(1) I’m an activist. I know people that work in the Nationals. It is more fun than phone, or visiting them at home.I’m followed by many(2) Valuable Internet Marketing is not kids & non-proﬁts? The huﬀ stole from me several times. They & twitter use an app on my acct? (3) Shaw is eliminate Prejudice, & a HR abuser! Also Up-search is linked to Philadelphia Scientology BDSM Kare Krishna, IRA bestiality(4) International Pedophile ring;(((((5)

Michel A. Rathier (@MARathier), 30 Mar Hi Diana, in answer to your Q..so as to amplify the message and contribute to increasing its broadcast

Samiul Hoque (@samhq7), 30 Mar I want to see media spreading right news, not spreading lies. So, I tweet real facts and tag

86 media to let them spread the truth.

brotherjohn73 (@brotherjohn73), 30 Mar it amazes me how we let them have control, it is important to me to let them know I am awake. Freedom is my emotion.(1) check out my list. There are descriptions on diﬀerent tags I use.(2)

Mike beadow (@Mwindtalker), 30 Mar I do not tweet 2 Media outlets,I Tweet AT them..Mainstream Media is Owned by Global (cont) http://tl.gd/lefv82 [ Corporations & do not present a balanced view of events happening around the world..The editors of these Media Outlets twist the News 2 suit the Corp. Agenda.. Here is a perfect example of how far we have 2 go in order 2 Grab the attention of people.. http://www.youtube.com/watch?feature=player detailpage&v=nA-z3ytS5nE ..It took a major protest in front of CNN’s Building in LA B4 Major Media revealed what is happening..That is why I Tweet @ Mainstream Media so individuals such as Urself C it...

Andrea( @AGraceMorgan63), 30 Mar just a few of the news papers and outlets I read.

Darren Gudmundson (@dgudmundson), 30 Mar Having a target or recipient gives meaning to the comment. Gets picked up in search. And, yes.

Abbas (@Abbas Abedi), 30 Mar They provide a soundboard for my views.

Mehdi (@MehdiMasud), 30 Mar I did it so dat they can’ve d interest to observe this fact. Cz in #BD injustice &violating humanity going on in d name of justice!

Fahmim Ferdous (@fahmimF), 30 Mar to let them know what’s happening in my country; to try & force them to notice what they chose to conveniently ignore.

Jim Fowler (@Jim Fowler1), 30 Mar You’re welcome. As to what I would change about media agenda, they too often follow

87 political elites’ agenda instead of looking...(1) ...for more substantive stories and trying to ﬁnd out the truth, rather than he said/she said, horse race kind of coverage. (2) Just to clarify, I’m a big believer in the necessity of professional journalism, just frustrated with its direction.(3)

Brad Orton (@BradANGSA), 30 Mar I bring interesting stories to the attention of the other media outlets...they use some of them as ﬁll-ins...

Paul Hoyt( @paulhoyt911), 30 Mar it allows people to understand that there are things out there and people may need to get their ideas and points of view to them.

Thomas Silvey (@formula1uberfan), 30 Mar are you talking media like @FoxNews or @CBSNews or @CNN? If so, the reason I do so is 4 to tweet what I think about a subject...

Jason Jusino (@Jtitoj01), 30 Mar to keep them aware that OperationMockingBird is/was successful and Alternative Media is light years ahead. Basically, We are awake.

Linda Wink (@winkelle,) 30 Mar tweeting to media is comparable to a message in a bottle.

Pierce Gibson Crosby (@CrosbyVenture), 30 Ma r For : The 3 ”channels” tagged represent emerging market pioneers who r competing 4 new audiences. I tag them to direct the argument. aSouthernLass (@aSouthernLass), 29 Mar Tagged all the networks to grab their attention. Out of all these ONE had the story of our president-turned-lobbyist. Disgraceful.

Vitaly Kroy (@VitalyKroy), 29 Mar I tweet them because they seem to lack backbone to ask the tough questions because America fell out of touch w/ politics.

88 Zark Riggs (@ZarkRiggs), 29 Mar I don’t know how it is in Sweden but here, some of the networks, one in particular, are only interested in propagandism.

HEADLINE ZOO (@HeadlineZoo), 29 Mar Visibility(1) #BjornBorg was an incredible athlete but as #StanSmith pointed out he wore cleats @ #Wimbledon which was prohibited. Your thoughts?(2)

Tracy Watson( @trayce), 29 Mar that was a rt bec they guy who tweeted it was an idiot.

Jon H (@Merlin333), 29 Mar various reasons. 1.) The on-air talent rarely is in charge of content,often an associate or mgmnt reviews social media remarks (1) 2.) Because I hear something I want to respond to 3.) To express my feelings pro or con regarding the story,talent,info etc(2) 4.) To be sure a record is created in case what I consider to be outrageous content e.g. When @cnn ’s @johnking suggested the @gop (3) 5.) In case I want to be sure comment gets to more that one person e.g. @JoeNBC is known to block detractors.(More but working) (4) adopt Lee Atwater tactics, tactics even those w/ cursory knowledge of these tactics know to be named the ”nigger plan”!(5) What ”I’d like to change most” is to have reporters having staﬀ assisting knowing more than me,unafraid to challenge guests(6)

TheLeader2013 (@ImGhettoMessiah), 29 Mar if they ever check their mentions...its just so they know that ppl are holding them accountable for their ﬁctitious news reporting

Marty Left (@MartyLeft), 29 Mar i am trying to reach as many ppl as possible show a diﬀerent view point.

Saturn del Strega (@cosmic escape ,) 29 Mar why not?

89 Cean (@ARTCETA), 29 Mar My tweets (2 #M$M ) are unvarnished, objective #truth #Reporter-At-Large #TruthSeekers #Fair&Square #UnzooObfuscation #No$Agenda

Patriot Boots (@m2guru), 29 Mar I replied to someone else who had included them. Typically I only include my heroes if they RT me, or if I’m quoting them. #????? (1) In particular, those are some of the most well known #Conservative thinkers in America. (And most despised by the Left.) #tcot(2)

Abbas Islam Khan (@abbasnoman), 29 Mar Aware International Media!

AAmom (@AVD911), 29 Mar I assume they hired some intern to read everything tweeted about them (collecting data). The @meetthepress boycott is still hot!

Cean (@ARTCETA), 29 Mar #M$M isn’t ”reporting” accurately bc ratings, bias & agenda POV. #NeutralReporting #TruthIsBest vs tabloid commentary w/agenda

Soldier Rojob Ali (@rojob ali), 28 Mar because this conect me with al the greatest people of the world.I can express my voice waht i need to inform to the outer world.

Guto (@augustoafm,) 28 Mar Your question it’s about my tweets ”indigenous of the Aldeia Maracan”?

The Clothing Line (@RobertoRaymon), 28 Mar because we know they will ignore it.

Lloyd Wells (@lloydbgwells), 27 Mar To let them know that we know they are ignoring news that eﬀects the lives of ppl and America. And that they are bias. Sheep.

90 Sue Smith (@SueSmith511), 27 Mar I see hypocrisy in some news outlets. When I do, I call them out & try to show another side to those who blindly follow

Sue Sen (@sosforanimals), 27 Mar Media picks up a lot of the time & they run stories right soon after. You will see this with NBC,NYT etc.They see your tweets!

Khurram Qayam (@kqbrain), 27 Mar I try to approach wider audience who r generally b told only some part of story. The global village need to b understood by commons

Larry Richman (@larry411), 27 Mar I don’t understand your questiojn.

PAUL X BROW (@guardbodX), 27 Mar It’s a way to communicate! make friends and be more along with people from other countries #quote #PXbrow god bless(1) Yes! people for years had no interaction channel as powerful as Twitter to shout out loud or to assert their rights #quote (2) The twitter is a media response quickly! write to justin bieber now or for barack obama in the certainty that they will read(3) one of my ﬁghts is against pedophilia and sex tourism read more http://paulxbrow.blogspot.com.br /2013/02/sells-up-pieces-of-meat-from-bodies-of.html @cnn @nytimes (4) Even if people do not respond they will read and know your opinions! this is internet (twitter before / after twitter) #PXbrow hugs(5)

Enigma77 (), 27 Mar I feel strongly that the corporate owned media is nothing more than a tool to keep the public uninformed or even misinformed.

Harvey Smith (@Harvey1966), 27 Mar What do you mean?

Jim Fowler (@Jim Fowler1), 25 Mar Western media isn’t giving much coverage to what is happening in Burma, so I added a few

91 media outlets. (1) Seems like media excuse for not being more substantive is it isn’t good for business so I wanted 2 show them people care about this(2) I picked CNN, MSNBC, & NYT b/c they’re prominent, reputable outlets. If I had more characters, would’ve added WaPo, LATimes, etc.(3) I didn’t bother with Fox News, because I doubt they or their viewers are interested. (4) If Muslims were the attackers, instead of victims, Fox would probably be all over it. Let me know if you have other Qs. Glad 2 help(5)

Shahzeb Suleman (@Shahzeb9968), 25 Mar I want media people to get ideas against paksitan’s politics in last ﬁve years ;) yasooyasoo (@yasooyasoo2012), 25 Mar Thanks Diana! We jpn twittists(?) want to tell our infomation to friends abroad. I want to be Palesitinians’ real friends. @cnnbrk

Eric Parks (@erictparks), 24 Mar I see CNN as the US mainstream channel of record on intl news.BBC is unparalleled in news culture. Subjective but there you have it

NO! GOP! NO! (@jeanepeters13), 24 Mar I expect nothing from Fox. They will lie all day long. Nothing too far fetched!! But...CNN..I’m disappointed.

Grant Petersen (@GP 011), 24 Mar sure. Reason being main stream media seems to turn a blind eye to world atrocities if whites or Christians are not dying.